Accidental foreign body (FB) inhalation is a life-threatening event that occurs most frequently in children aged 3 yr or less and has a reported mortality rate between 0% and 1.8%.1–4 In China, the incidence is higher because of feeding habits, such as the popular snacks including plant seeds. Our hospital is an ear-nose-throat specialty institution in southern China that manages nearly 200 cases of airway FB removal each year.
An inhaled FB may lodge in different sections of the bronchus and cause pneumonia, whole or segmental atelectasis, and/or hyperinflation distal to the obstruction. Prompt and early removal of a FB with a rigid bronchoscope is essential for reducing complications and mortality rates.5,6 In children, bronchoscopy for FB removal is performed under general anesthesia; however, the use of spontaneous ventilation (SV) or controlled ventilation (CV) varies among anesthesiologists.
Clinically, hypoxemia is the most commonly observed adverse event in rigid bronchoscopy. The causes of hypoxemia include improper manipulation of the surgical instruments in the airway, partial or complete airway obstruction by the FB, and inflammation secondary to FB.7,8 In addition, a coexisting medical condition, surgical and anesthetic techniques can also affect hypoxemia. Zaytoun et al.9 reviewed the records of 504 patients during a 10-yr period and demonstrated that the most important variables in predicting the occurrence of complications were history of bronchoscopy, duration of the procedure, and type of FB. However, ventilation mode and anesthetic techniques were not discussed.
Therefore, we designed a prospective nonrandomized observational study to encompass all children younger than 5-yr-of-age who underwent rigid bronchoscopy for FB removal during a 2-yr period. The purpose of the study was to investigate risk factors that statistically correlated with intraoperative or postoperative hypoxemia.
Approval was obtained from the hospital's Human Research Committee before initiating the study. Between January 2007 and December 2008, 384 children younger than 5 yr underwent rigid bronchoscopy for suspicion of FB aspiration. All surgical manipulations were performed with Karl-Storz rigid bronchoscopes under general anesthesia. Informed consent was obtained from parents or legal guardians before the anesthetic and surgical procedure. Clinical information such as age, gender, clinical signs and symptoms, FB duration, FB type, and FB location were collected. In addition, chest radiography findings, anesthetic technique, intraoperative ventilation mode, duration of procedure, and duration of emergence from anesthesia were also recorded. Perioperative adverse events including hypoxemia (pulse oxygen saturation below 90%), breath holding, laryngospasm (stridor on inspiration with closure of vocal cords), postoperative airway obstruction, pneumothorax, and the need for reintubation were recorded. Children were excluded from the study if they had preexisting pneumothorax, mediastinal emphysema, apnea, or a cardiopulmonary arrest. If the first bronchoscopic procedure to retrieve the FB was unsuccessful, a second attempt was made 3–5 days later. Patients who had a thoracotomy to remove the FB after two failed bronchoscopic procedures were also excluded from the study.
Data from all patients were divided into two groups based on the ventilation technique used for the surgical procedure: the CV group, in which the patient received a muscle relaxant; and the SV group. In the CV group, children were further divided into two subgroups. One group received manual intermittent positive pressure ventilation (MPPV), and the other group was ventilated with manual jet ventilation (MJV). Patients in the SV group were also divided into two subgroups. One group received total IV anesthesia (TIVA), whereas the other group of patients were anesthetized with sevoflurane. The ventilation mode and anesthetic technique were chosen based on the anesthesiologist's preference, and in some cases, on the surgeon's preference.
Before the start of the anesthetic, all children were premedicated with atropine (0.01 mg/kg) and corticosteroids. In the MPPV and MJV groups, anesthesia was induced with propofol (3–5 mg/kg), remifentanil (1 mcg/kg), and succinylcholine (2 mg/kg). Anesthesia was maintained by infusion of propofol (100–150 μg/kg/min) and remifentanil (0.1 μg/kg/min), with bolus doses of succinylcholine (1 mg/kg) as clinically indicated. Anesthetic induction and maintenance in the TIVA group was the same as in the MPPV and MJV groups except for the exclusion of succinylcholine. In the sevoflurane group, mask induction was performed with 8% sevoflurane carried by 8 L/min oxygen flow and anesthesia was maintained with 3%–4% sevoflurane and 4 L/min oxygen flow. All children received topical 1% lidocaine (3–5 mg/kg) sprayed on the surface of the vocal cords and trachea.
After insertion of a rigid bronchoscope, the respiratory circuit was connected to the side port of the bronchoscope in the MPPV, TIVA, and sevoflurane groups to sustain adequate ventilation and oxygenation. In the MJV group, a modified small suction catheter (internal diameter = 1.5 mm, with the lateral aperture sealed) connected to a manually controlled jet ventilator (VBM, Germany) was inserted through the nasal passage into the trachea, and oxygen was delivered with a driving pressure of 15–35 psi (usually <25 psi) at a frequency of 16–20 bpm. The effectiveness of MJV was assessed by degree of chest excursion. Assisted ventilation over spontaneous respiration in the TIVA and the sevoflurane groups was performed at times to achieve adequate ventilation. Heart rate, heart rhythm, and oxygen saturation were continuously monitored during the procedure using electrocardiogram and pulse oximetry. Noninvasive arterial blood pressure was measured at 5-min intervals. Once the FB was removed, assisted ventilation or spontaneous respiration was maintained through a facemask in the MPPV, TIVA, and sevoflurane groups. For the MJV group, MJV was continued until spontaneous respiration resumed.
All data were analyzed by SPSS 11.5 software. The variables with normal distribution were expressed as mean ± sd (mean ± sd), and Student's t-test was used for comparison between those groups. Data with a skewed distribution were presented as median (interquartile), M(Q), and Mann–Whitney U-test was used for comparison of variables between corresponding groups. The numeration data were expressed as numbers or percentages, and the comparisons between groups were performed with Pearson x2-test. A P value <0.05 was considered statistically significant.
An amalgamated dataset was used for the analysis of factors correlated with perioperative hypoxemia. The event of hypoxemia was the main outcome. Variables considered in the multivariate analysis for intraoperative hypoxemia included gender, age in months (divided into <12, 12–24, 25–36, 37–48, >48 mo), weight, duration that FB was retained, type of FB, premedication, pneumonia before operation, body movement when the bronchoscope was inserted, duration of operation, surgeon's experience, extent of airway obstruction, FB location, and ventilation mode (SV, MPPV, or MJV). Additionally, duration of emergence from anesthesia after withdrawal of the bronchoscope and postoperative airway obstruction were included as variables when assessing risks for postoperative hypoxemia. Patients with no FB identified were excluded in this dataset. All variables with a P value <0.2 in single-factor analysis were considered for further multivariate analysis using a logistic regression model. A P value <0.05 was considered statistically significant.
Over a 2-yr period, 384 patients met the inclusion criteria for the study and underwent rigid bronchoscopy. Among them, 327 had successful FB removal (85%) and 57 (15%) had no FB identified intraoperatively. All surgical manipulations and anesthetic management were performed by skilled and experienced staff who had performed at least 50 FB removal cases before. The clinical characteristics of all groups of children are listed in Table 1. Distribution of age, gender, weight, duration that FB was retained, and type of FB were not significantly different among the four anesthetic technique groups. There were more male than female patients and more organic than inorganic FBs in each of the groups. The results of surgery are listed in Table 2. The duration of the procedure was not significantly different among the four anesthetic technique groups. However, significantly longer duration of emergence from anesthesia and significantly lower percentage of successful FB removal was observed in the TIVA group compared with any of other three groups. Three patients required thoracotomy to remove the FB, which were not included in the analysis. Among them, one was in the TIVA group, one was in the MJV group, and one was in the MPPV group. The types of FB for these three patients were a peanut, a pen cap, and a toy part, respectively. Almost half of the FBs were located in the right bronchus (49%), about one third in the left bronchus (38%), and the remainder in the trachea (13%). The distribution of FB location among the four anesthetic technique groups was not significantly different.
The observed perioperative adverse events of intraoperative body movement, hypoxemia, breath holding, postoperative laryngospasm, and airway obstruction were prevalent in all groups of patients (Table 3). Children in the TIVA group showed more intraoperative body movement, breath holding, and postoperative laryngospasm than those in the other groups. Children in the MJV group presented with the least incidence of intraoperative hypoxemia of all four groups. Pneumothorax was the most severe complication, with nine cases confirmed by radiography. Three of these patients were in the MPPV (2.9%), four in the MJV (3.1%), and two in the TIVA group (2.4%). However, the difference among groups was not statistically significant (P = 0.550, using Fisher's exact test). All patients who developed pneumothorax required chest drainage through the injured side for 4–6 days. This resulted in longer hospital length of stay of 6–8 days compared with 2–3 days for patients without a pneumothorax. Pneumothorax resolved in all patients without sequelae. Reintubation due to hypoxemia at the end of bronchoscopy was rare and did not occur in any of the MJV group patients. Changing of ventilation mode intraoperatively was noted only in the shifting from SV to CV. There were no deaths in any of the groups.
According to the results from single-factor analysis, variables screened out for further multivariate analysis for intraoperative hypoxemia included age (P = 0.039), type of FB (P = 0.025), pneumonia before operation (P = 0.067), intraoperative body movement (P = 0.011), duration of operation (P = 0.044), extent of airway obstruction (P = 0.104), and ventilation mode (P < 0.001). The final logistic regression model is shown in Table 4. Factors that strongly correlated with intraoperative hypoxemia included age, type of FB, duration of surgical procedure, pneumonia before procedure, and ventilation mode. The risk of intraoperative hypoxemia increased as age decreased (P = 0.048, odds ratio [OR] = 0.668). No correlation was found between hypoxemia and the presence of an organic FB versus inorganic FB. When the types of plant seed were listed individually, the risk of intraoperative hypoxemia increased significantly (P = 0.027, OR = 2.654). A longer duration of surgery (>20 min) and pneumonia before operation also increased the risk of intraoperative hypoxemia significantly (P < 0.001, OR = 1.150 for surgical duration; P = 0.003, OR = 3.837 for pneumonia). Compared with MPPV, SV increased the risk of hypoxemia, whereas MJV decreased it (P = 0.035, OR = 3.152 for SV; P < 0.001, OR = 0.161 for MJV).
Similarly, variables screened out for postoperative hypoxemia included age (P = 0.100), type of FB (P = 0.014), intraoperative body movement (P = 0.012), duration of emergence from anesthesia (P = 0.007), postoperative airway obstruction (P = 0.004), surgeon's experience (P = 0.017), and ventilation mode (P = 0.110). The final logistic regression model is shown in Table 5. Two factors were found to be closely correlated with postoperative hypoxemia: the type of FB (P = 0.036, OR = 3.289) and duration of emergence from anesthesia at the end of the procedure (P < 0.001, OR = 1.164).
The results of the current study indicated that FB inhalation was more common in males than females and with more organic substances than inorganic substances. FBs were generally lodged in the right bronchus. These results were in accordance with most of the published epidemiologic and demographic reports.10–12 We recorded details of a variety of moderately adverse complications and observed nine occurrences of pneumothorax, but noted no incidence of tracheobronchial bleeding. The overall complication rate was comparable with that reported in the literature.3,12,13
Intraoperative or postoperative hypoxemia was found to be the most frequent adverse event, and it could lead to a life-threatening outcome if not promptly treated. Its occurrence may depend on a variety of factors including the property of the FB, the surgeon's experience, anesthetic method, and patient's condition. In the current study, the surgeon's experience did not correlate with hypoxemia. This may be explained by the fact that our institution is an ear-nose-throat specialty hospital with a large number of admissions of pediatric patients with FB aspiration each year, and that all rigid bronchoscopies in this study were performed by very experienced surgeons.
Our study identified five factors that strongly correlated with increased intraoperative hypoxemia: age of patient, plant seeds as FB, pneumonia before procedure, long surgical duration, and ventilation mode. These factors could also have an impact on each another. The risk of hypoxemia increases in younger patients because of a more rapid oxygen desaturation as a result of lower functional reserve capacity and higher oxygen consumption. Plant seeds lodged in the bronchi may lead to a chemical reaction from released arachidonic acid-like substances and cause pneumonia.14,15 As a result, airway swelling and increased secretions make visualization of the FB difficult, make patients more prone to bleeding, prolong the duration of operation, and increase the occurrence of bronchospasm when manipulating the surrounding inflamed mucosa.
We also noted that the incidence of postoperative hypoxemia was related to duration of emergence from anesthesia and to the type of FB. The explanation for this phenomenon may be that long duration of emergence from anesthesia is related to deep anesthesia at the end of bronchoscopy. Decreased respiratory drive makes it harder to expectorate secretions from inflamed bronchi. Bucking, coughing, or laryngospasm may be induced by secretions accumulated under the glottis during the emergence period. Our finding was in accordance with a previous report by Blair et al.16 and partly in agreement with Tait et al.17
Of all the factors related to hypoxemia or other complications in this study, ventilation mode was strongly associated with intraoperative hypoxemia. An extensive search of Medline reports revealed that the optimal ventilation mode during rigid bronchoscopy for FB removal is still actively debated. In general, SV was more popular and was advocated before the mid 1990s,18–21 whereas more recently, reports in favor of CV have appeared.22,23 In a review by Farrell,24 the advantages and disadvantages of SV and CV were discussed; however, no personal preference was suggested.
Jet ventilation in rigid bronchoscopy was first introduced by Sanders in 1967.25 Since then, it has been modified and widely used in suspended laryngoscopy.26,27 Its use in pediatric FB removal has not been widely advocated. In this study, we developed a MJV method in which a small catheter was placed transnasally into the trachea for oxygen delivery. This technique produced fewer episodes of intraoperative hypoxemia than any other ventilation mode. The value of MJV became more pronounced when the bronchoscope had to be inserted distal to the FB. In this instance, ventilation became limited to the lung distal to the FB. In this situation, MJV became advantageous over the other ventilation modes, because MJV provided continuous ventilation in the noninvolved lung with a catheter separated from the bronchoscope. Thus, the risk of hypoxemia is reduced dramatically with MJV. The most serious complication of MJV is barotrauma.28 In our study, there were four confirmed cases of pneumothorax in the MJV group, but the incidence was not significantly different when compared with the other groups (three patients in the MPPV group and two in SV group). Further investigation found that all those patients had a FB retained for longer than 1 mo and required prolonged operative time. We found no evidence that correlate jet ventilation with pneumothorax.
Although perioperative adverse events were present across all four anesthetic technique groups, these complications seemed more prevalent in children who received the TIVA technique. Our study found increased intraoperative body movement, breath holding, and postoperative laryngospasm in the TIVA group. An inadequate depth of anesthesia could be one of the causes of these complications. The reason for the relatively frequent incidence of postoperative laryngospasm may have been due to depressed airway reflex during the considerably prolonged emergence stage associated with TIVA. The reason for the higher failure rate for FB removal in the TIVA group was unclear and may have been partly due to nonideal surgical conditions.
In conclusion, hypoxemia is the most frequent adverse event in rigid bronchoscopy for FB removal. The risk factors that correlated with intraoperative hypoxemia included patient age, type of FB, duration of surgical procedure, pneumonia before the procedure, and ventilation mode. Use of SV increases the risk of hypoxemia, whereas MJV decreases the incidence. For postoperative hypoxemia, our study observed that the nature of the FB and the duration of emergence from anesthesia after bronchoscopy are significant risk factors.
The authors acknowledge Dr. Fay Jou, an anesthesiologist in Cincinnati Children's Hospital Medical Center, for the delicate polish of English language.
1. Banerjee A, Rao KS, Khanna SK, Narayanan PS, Gupta BK, Sekar JC, Retnam CR, Nachiappan M. Laryngo-tracheo-bronchial foreign bodies in children. J Laryngol Otol 1988;102:1029–32
2. Mu L, He P, Sun D. Inhalation of foreign bodies in Chinese children: a review of 400 cases. Laryngoscope 1991;101:657–60
3. Hasdiraz L, Oguzkaya F, Bilgin M, Bicer C. Complications of bronchoscopy for foreign body removal: experience in 1,035 cases. Ann Saudi Med 2006;26:283–7
4. Aytaç A, Yurdakul Y, Ikizler C, Olga R, Saylam A. Inhalation of foreign bodies in children. Report of 500 cases. J Thorac Cardiovasc Surg 1977;74:145–51
5. Swanson K. Airway foreign bodies: what's new? Semin Respir Crit Care Med 2004;25:405–11
6. Weissberg D, Schwartz I. Foreign bodies in the tracheobronchial tree. Chest 1987;91:730–3
7. Litman RS, Ponnuri J, Trogan I. Anesthesia for tracheal or bronchial foreign body removal in children: an analysis of ninety-four cases. Anesth Analg 2000;91:1389–91
8. Tomaske M, Gerber AC, Weiss M. Anesthesia and periinterventional morbidity of rigid bronchoscopy for tracheobronchial foreign body diagnosis and removal. Paediatr Anaesth 2006; 16:123–9
9. Zaytoun GM, Rouadi PW, Baki DH. Endoscopic management of foreign bodies in the tracheobronchial tree: predictive factors for complications. Otolaryngol Head Neck Surg 2000;123:311–6
10. Surka AE, Chin R, Conforti J. Bronchoscopic myths and legends airway foreign bodies. Clin Pulm Med 2006;13:209–11
11. Hui H, Na L, Zhijun CJ, Fugao ZG, Yan S, Niankai ZK, Jingjing CJ. Therapeutic experience from 1428 patients with pediatric tracheobronchial foreign body. J Pediatr Surg 2008;43:718–21
12. Jose RG, Salvato F, Paula M, Ibraimo M, Anto B. Tracheobronchial foreign bodies in children: fifteen years of experience in a respiratory endoscopy unit. J Bronchol 2007;14:79–82
13. Pinzoni F, Boniotti C, Molinaro SM, Baraldi A, Berlucchi M. Inhaled foreign bodies in pediatric patients: review of personal experience. Int J Pediatr Otolaryngol 2007;71:1897–903
14. Svedström E, Puhakka H, Kero P. How accurate is chest radiography in the diagnosis of tracheobronchial foreign bodies in children? Pediatr Radiol 1989;19:520–2
15. Chatterji S, Chatterji P. The management of foreign bodies in air passages. Anaesthesia 1972;27:390–5
16. Blair JM, Hill DA, Bali IM, Fee JP. Tracheal intubating conditions after induction with sevoflurane 8% in children: a comparison of two intravenous techniques. Anaesthesia 2000;55: 774–8
17. Tait AR, Malviya S, Voepel-Lewis T, Munro HM, Seiwert M, Pandit UA. Risk factors for perioperative adverse respiratory events in children with upper respiratory tract infections. Anesthesiology 2001;95:299–306
18. Martinot A, Closset M, Marquette CH, Hue V, Deschildre A, Ramon P, Remy J, Leclerc F. Indications for flexible versus rigid bronchoscopy in children with suspected foreign-body aspiration. Am J Respir Crit Care Med 1997;155:1676–9
19. Inglis AF Jr, Wagner DV. Lower complication rates associated with bronchial foreign bodies over the last 20 years. Ann Otol Rhinol Laryngol 1992;101:61–6
20. Meretoja OA, Taivainen T, Räihä L, Korpela R, Wirtavuori K. Sevoflurane-nitrous oxide or halothane-nitrous oxide for paediatric bronchoscopy and gastroscopy. Br J Anaesth 1996;76: 767–71
21. Perrin G, Colt HG, Martin C, Mak MA, Dumon JF, Gouin F. Safety of interventional rigid bronchoscopy using intravenous anaesthesia and spontaneous assisted ventilation, a prospective study. Chest 1992;102:1526–30
22. Metrangolo S, Monetti C, Meneghini L. Eight years' experience with foreign-body aspiration in children: what is really important for a timely diagnosis. J Pediatr Surg 1999;34:1229–31
23. Soodan A, Pawar D, Subramanium R. Anaesthesia for removal of inhaled foreign bodies in children. Pediatr Anaesth 2004;14:947–52
24. Farrell PT. Rigid bronchoscopy for foreign body removal: anaesthesia and ventilation. Pediatr Anesth 2004;14:84–9
25. Sanders RD. Two ventilating attachments for bronchoscopes. Del Med J 1967;192:170–5
26. Rezaie-Majd A, Bigenzahn W, Denk DM, Burian M, Kornfehl J, Grasl MCh, Ihra G, Aloy A. Superimposed high-frequency jet ventilation (SHFJV) for endoscopic laryngotracheal surgery in more than 1500 patients. Br J Anaesth 2006;96:650–9
27. Bourgain JL, Desruennes E, Fischler M, Ravussin P. Transtracheal high frequency jet ventilation for endoscopic airway surgery: a multicenter study. Br J Anaesth 2001;87:870–5
© 2009 International Anesthesia Research Society
28. Baer G. Complications and technical aspects of jet ventilation for endolaryngeal procedures. Acta Anaesthesiol Scand 2000;44: 1273–4