Aspiration of foreign material into the airway can result

Transcription

1 J Vet Intern Med 2010;24: The Role of Bronchoscopy in Foreign Body Removal in Dogs and Cats: 37 Cases ( ) A.C. Tenwolde, L.R. Johnson, G.B. Hunt, W. Vernau, and A.L. Zwingenberger Background: Foreign body aspiration is a differential diagnosis for acute or chronic cough that requires medical or surgical management in animals. Hypothesis: Success of bronchoscopy in airway foreign body removal is dependent on the size of the animal, duration of clinical signs, and location of the foreign body. Animals: Thirty-two dogs and 5 cats with airway foreign bodies identified at the UC Davis Veterinary Medical Teaching Hospital. Methods: Retrospective case study evaluating the role of duration of clinical signs and body size in successful bronchoscopic removal of foreign bodies. In addition, radiographic localization of disease was compared with bronchoscopic identification. Bronchoalveolar lavage (BAL) culture and cytologic findings are reported. Results: Bronchoscopy was successful for removal of airway foreign bodies in 76% of animals (24/28 dogs and 2/5 cats), and in dogs was independent of duration of clinical signs or body size. One-third of thoracic radiographs lacked distinctive features of an airway foreign body, and therefore radiography was unable to predict the affected site. BAL fluid at the site of the foreign body contained more neutrophils and more often had intracellular bacteria than lavage fluid from a separate site. Conclusions and Clinical Importance: Bronchoscopy was successful in removing airway foreign bodies regardless of animal size or long duration of clinical signs. Results of this study confirm the utility of bronchoscopy with lavage in management of suspected foreign bodies, even in the absence of localizing radiographic findings. Key words: Cytology; Microbiology; Respiratory endoscopy; Thoracic radiography. Aspiration of foreign material into the airway can result in acute or chronic cough and respiratory distress. Depending on the type of foreign material aspirated, the degree of airway obstruction, and the consequent infection and inflammation at the site, various imaging modalities can be useful in diagnosing and localizing disease associated with airway foreign bodies. A recent study reported that the most common radiographic abnormalities of focal interstitial to alveolar infiltrates could be identified in approximately 2/3 of cases. 1 The study suggested that computed tomography (CT) might be more accurate in identifying foreign body location before interventional therapy, but this imaging modality is not as widely available as is survey radiography. Some foreign bodies remain in the airways whereas others migrate through the lung parenchyma and cause lobar consolidation, pneumothorax, or pyothorax. Depending on the site of disease, bronchoscopy or fluoroscopy can be a noninvasive method for management of disease in dogs and cats. 2,3 When these procedures are unsuccessful or when pneumothorax or pyothorax are present, surgical intervention often is required. From the William R. Pritchard Veterinary Medical Teaching Hospital (Tenwolde, Johnson, Hunt, Vernau, Zwingenberger), the Department of Medicine & Epidemiology (Johnson), the Department of Surgical and Radiological Sciences (Hunt, Zwingenberger), and the Department of Pathology, Microbiology and Immunology (Vernau), University of California School of Veterinary Medicine, Davis, CA. Corresponding author: Lynelle Johnson, William R. Pritchard Veterinary Medical Teaching Hospital, University of California School of Veterinary Medicine, Davis, CA; lrjohnson@ucdavis.edu. Submitted April 27, 2010; Revised June 6, 2010; Accepted June 25, Copyright r 2010 by the American College of Veterinary Internal Medicine /j x Abbreviations: BAL CT UCD-VMTH bronchoalveolar lavage computed tomography University of California, Davis Veterinary Medical Teaching Hospital Migrating grass awns (ie, foxtails) are common in northern California. We hypothesized that bronchoscopy would be less successful for removing airway foreign material in smaller animals because of their smaller airway diameter resulting in a more technically difficult procedure. We also presumed that bronchoscopy would be less successful in animals with longer duration of clinical signs. In this study, we reassessed the ability of radiographs to accurately predict the location of airway foreign bodies based on previously identified characteristics. Finally, we hypothesized that bronchoalveolar cytology of fluid collected from the site of a foreign body would differ from that obtained at an alternate site and would be more likely to indicate septic inflammation. Materials and Methods This study was performed as a retrospective review of case records at the University of California, Davis Veterinary Medical Teaching Hospital (UCD-VMTH). Medical records from were searched for a diagnosis of a pulmonary foreign body in dogs or cats. Search fields included the visit summary, diagnostic procedures, endoscopy reports, and histopathology records for reference to airway or pulmonary foreign body. Cases were excluded if only the mediastinum was involved or if a foreign body was not found within the airways or lung parenchyma either during the endoscopic procedure or during histologic examination of resected tissue. Animals that did not have diagnostic testing performed but that had a foreign body discovered at necropsy also were excluded

2 1064 Tenwolde et al Fig 1. Lateral (A) and dorsoventral (B) radiographs of a dog with a foreign body in the left cranial lung lobe. Radiographs were scored for the level of certainty that imaging characteristics represented foreign body aspiration with 1 5 questionable, 2 5 possible, 3 5 probable, 4 5 most likely, 5 5 definite. There were no localizing signs for the foreign body and radiographs were assigned a score of 1. from analysis. Medical records were reviewed for duration of clinical signs (including cough, labored breathing, or respiratory distress), results of endoscopic or surgical recovery of foreign material, bronchoalveolar lavage (BAL) cytology, and culture findings. Thoracic radiographs were reviewed for the type, severity, and location of pulmonary infiltrates as well as the presence or absence of pleural disease by a board-certified radiologist (A.L.Z.) who was aware of the diagnosis but was blinded to the outcome of the case. Radiographs lacking pulmonary infiltrates were identified. The level of certainty in the radiographic diagnosis of foreign body as assessed by characteristics previously defined was graded as 1 5 questionable, 2 5 possible, 3 5 probable, 4 5 most likely, 5 5 definite (Fig 1). Level of certainty in the location of foreign material was assessed as 1 5 questionable, 2 5 possible, 3 5 probable, 4 5 most likely, 5 5 definite for the region considered most abnormal on thoracic radiographs (Fig 2). 1 The diagnostic evaluation of individual cases was directed by the admitting clinician, and consultation with faculty and other clinical services was sought as needed. A minimum database was collected and analyzed in all animals. In various cases, thoracic ultrasound examination, CT, bronchoscopy, exploratory thoracotomy, or some combination of these procedures was performed to investigate presenting complaints of cough or respiratory difficulty. Animals undergoing bronchoscopy were preoxygenated and general anesthesia was induced by standard protocols. For animals large enough to accommodate a size 7 endotracheal tube, the endoscope was passed through a special T-adapter and gas anesthesia was employed. In smaller animals, bronchoscopy was performed with a propofol infusion at mg/kg/min and oxygenation was maintained by jet ventilation administering 180 breaths/min. Pulse oximetry, EKG, and blood pressure were monitored throughout the procedure in all animals. The bronchoscopes available for use included 2 small endoscopes with a 1.2 mm biopsy channel for instruments 1.0 mm in size, a and 4 larger endoscopes with a 2.0 mm biopsy channel for instruments 1.8 mm in size. b Endoscopic equipment available for foreign body retrieval included biopsy forceps, alligator forceps, 3 prong grabbers, and a snare. After identification and removal of a foreign body, the endoscope was withdrawn from the airway, the instrument channel was flushed with sterile saline, and the exterior was wiped with saline soaked gauze sponges. BAL was performed at the site of the foreign body in most cases, and in some cases a separate site distant from the foreign body site was also lavaged at the discretion of the endoscopist. To obtain BAL fluid, an aliquot of warmed, sterile 0.9% saline was instilled through the biopsy channel of the endoscope, the channel was flushed with a small volume of air, and hand suction was applied to recover fluid that had been in contact with the bronchoalveolar space. Individual aliquot volume for BAL, number of aliquots instilled, and total volume utilized were determined by the endoscopist. Location, volume of fluid instilled, and volume recovered were recorded for each site. BAL fluid was submitted for cytology and microbiologic culture. Distinct sites within the lung Fig 2. Lateral (A) and dorsoventral (B) radiographs of a dog with plant foreign bodies in the left caudal lung lobe, Level of certainty in the location of foreign material for radiographs was assessed as 1 5 questionable, 2 5 possible, 3 5 probable, 4 5 most likely, 5 5 definite for the region considered most abnormal on thoracic radiographs. This case was assigned a score of 5 for the likelihood that a foreign body was present and a 5 indicating a high level of certainty for the location of the foreign body.

3 Bronchoscopy for Foreign Body Removal 1065 Fig 3. Bronchoscopic image of a multipronged grass awn (83 cm) wedged in the right middle lobar bronchus in a dog from this study (A), and a mucus-coated branch from a thorny bush (60.5 cm) obstructing a ventral branch in the left caudal lung lobe (B) in a 2nd dog. were analyzed separately for determination of total cell count and differential cell count based on a count of 200 cells. Reported reference intervals for BAL cell and differential counts in dogs are cells/ml comprised of up to 5 7% eosinophils, neutrophils, or lymphocytes and 65 85% macrophages, whereas in the cat, 16 20% eosinophils is considered within normal limits. 4,5 The percentage of each type of inflammatory cell for each BAL was tabulated, and cytology reports were scrutinized for documentation of the presence of intracellular bacteria. A pooled BAL sample was submitted for aerobic, anaerobic, and Mycoplasma cultures. Surgical approach depended on the location of the foreign body as predicted by radiographs, thoracic CT, bronchoscopy, or some combination of these procedures. Either lateral thoracotomy or median sternotomy with lobectomy was performed at the discretion of the surgeon. Exact foreign body location was confirmed histopathologically for all surgically treated patients. Statistics Body weights of animals that had an airway foreign body successfully removed by bronchoscopy were compared with those that required surgery by an unpaired t-test. Duration of clinical signs was compared between the group of dogs with bronchoscopic removal of a foreign body and those that had surgery after bronchoscopy or surgery alone by a Kruskal-Wallis test for nonparametric data. The level of agreement between radiographic localization of the foreign body and bronchoscopic or surgical detection was assessed. BAL data are presented as median and range. Cell and differential counts were compared between the site of the foreign body and an alternate site within the lung by the Mann- Whitney test for nonparametric data. Because only small percentages of eosinophils and mast cells were detected, the number of samples containing eosinophils and mast cells was compared between sites with Fisher s exact test. Additionally, Fisher s exact test was used to compare the number of samples containing visible intracellular bacteria at the site of the foreign body and the alternate site. For all tests, P o.05 was considered significant. Results Between 2000 and 2008 at the UCD-VMTH, 37 medical records were identified with a diagnosis of a pulmonary foreign body in 32 dogs and 5 cats. Bronchoscopy was performed as the initial intervention in 28 of 32 dogs and 5 cats and was successful in retrieving airway foreign material in 24/28 dogs (86%) and 2/5 cats (40%). During this time period, 372 dogs and 138 cats had bronchoscopy performed, reflecting incidence rates of 8 and 4% in dogs and cats, respectively. Foreign bodies in dogs retrieved by bronchoscopy consisted of grass awns (19), other plant material (4), or plastic material (1) (Fig 3). In cats, small twigs were completely removed in 2/5 cases and plant material was partially removed bronchoscopically in 2 additional cats. In dogs treated successfully by bronchoscopy, 1 foreign body was removed in 19 dogs, whereas more than 1 foreign body was found in 5 dogs. In dogs, foreign bodies were removed by bronchoscopy from a right-sided bronchus in 11/24 (46%) cases, from a left-sided bronchus in 7/24 (29%) cases, from both right and left bronchi in 4/24 (17%) cases, and from the trachea or carina in 2/24 (8%) cases. Specific airways sites were carina (1), trachea (1), right middle bronchus (5), accessory bronchus (9), right caudal bronchus (5), left cranial bronchus (2), and left caudal bronchus (9). In 2 cats, material at the carina was successfully removed with the bronchoscope. In 21 dogs with successful removal of foreign material, endoscope use was recorded. In 3/21, a smaller bronchoscope a was employed and in the remaining 18, a larger bronchoscope b was employed. In 4 cats with recorded information, use of a small scope was successful in 1 case, whereas the remaining 3 procedures were unsuccessful despite the use of a larger endoscope. Surgery was performed as the first intervention in 4 dogs based on the radiographic presence of pneumothorax (1), consolidation of a lung lobe (1), CT findings of lobar consolidation (1), or identification of the foreign body on CT (1). Surgery followed unsuccessful bronchoscopy in 4 additional cases. Surgery, histopathology, or both confirmed foreign material in the right lung in 6/10 (60%) dogs and in the left lung in 2/10 dogs (20%). Bronchoscopy was followed by surgery in 4/28 (14%) dogs and 3/5 cats (60%). In dogs, bronchoscopy was unsuccessful because of pooled mucus or blood that obscured foreign material (n 5 3) or because the lesion was too distal to reach with instruments (n 5 1). In 3 cats managed surgically, foreign bodies were lodged at the left mainstem bronchus, accessory lobar bronchus, and right cranial lung lobe and could not be removed with instruments passed bronchoscopically. Two of these cats had more than 1 foreign body present, and lung lobectomy

4 1066 Tenwolde et al with successful foreign body retrieval was performed in both, 1 from the left cranial lobe and the other from the right accessory lobe. The 3rd cat with foreign material in the right cranial lobar bronchus also had a lobectomy performed. The most common location for a foreign body in cats of this study was within the trachea or carina. Overall, success was achieved by bronchoscopy alone in 26 cases, bronchoscopy followed by exploratory thoracotomy in 7 cases, and exploratory thoracotomy alone in 4 cases. Median age of dogs at the time of presentation was 3.5 years (range, 3 months to 12 years) and median age for cats was 7 years (range, years). Dog breeds represented more than once included Labrador Retriever (7), and 2 each of German Shepherd Dog, Jack Russell Terrier, Beagle, Rottweiler, and Vizsla. There were 20 male and 12 female dogs and 3 male and 2 female cats. Median body weight of dogs was 28.8 kg (range, 4 41 kg) and median body weight of cats was 4.2 kg (range, kg). Body weights of dogs in which bronchoscopy alone was successful (27.0 kg; range, 4 41 kg) did not differ significantly from those in which surgery was required (34 kg; range, kg) (P 5.22) and all dogs weighing o15 kg (n 5 4) had successful bronchoscopic removal of foreign material in this study. Duration of clinical signs in dogs successfully treated by bronchoscopy ranged from 0.5 days to 1 year (median, 21 days; mean, 91 days). In dogs that had bronchoscopy followed by surgery, duration of signs ranged from 14 to 90 days (median, 60 days; mean, 56 days), and in dogs treated initially with surgery, duration of signs ranged from 2 to 210 days. Duration of clinical signs did not differ significantly between dogs managed with bronchoscopy in comparison to those that required surgery (P 5.97). In 2 cats with successful removal of an airway foreign body by bronchoscopy, signs had been present for 1 hour and 14 days in comparison to 4 60 days in cats that had surgery performed. In all animals, clinical signs resolved with removal of the foreign body and appropriate antimicrobial treatment. Radiographs were available for review for 33 animals. Presence of a foreign body was correctly identified in 22/ 33 (66%) animals (most likely or definite, grades 4 and 5) and considered of low probability (questionable, possible, probable; grades 1 3) in 11/33 (33%) animals. Two of 11 (6%) low probability radiographs were considered probable, and 9/11 (27%) were considered questionable or possible. Of the 22 likely or definite radiographic classifications, 19 (86%) were correct for location. Eight of 9 (89%) radiographic examinations classified as questionable or possible were incorrect for location of foreign body, and in 7/9 (78%) the predominant radiographic finding was a generalized interstitial or bronchial pattern with no localizing features. Overall, the radiographic assessment of foreign body location was incorrect in 11/33 (33%) animals. BAL fluid was collected from 25 of 28 dogs that had bronchoscopy performed, however the site of lavage was not recorded in 4 dogs. In 21 dogs, lavage was performed at the site of the foreign body and in 17 dogs, an additional site was lavaged and analyzed. Total cell count did Table 1. Differential cell counts in bronchoalveolar lavage fluid collected at the site of the foreign body and at an alternate site within the lung. Site of Foreign Body Alternate Site P Value % Neutrophils 89 (4 100) 22.5 (2 89).004 % Lymphocytes 2 (1 17) 6.5 (1 20).007 % Eosinophils 0 (0 14) 1 (0 41).048 % Macrophages 10.5 (0 77) 57.5 (0 94).022 Data are presented as medians with range. not differ between the site of the foreign body (2,700/mL; range, ,500/mL) and an alternate site (1,229/mL; range, 200 4,100/mL) (P 5.068). However, fluid obtained from the lavage site had significantly more neutrophils and fewer macrophages, lymphocytes, and eosinophils than the alternate site (Table 1, Fig 4). The proportion of samples containing eosinophils or mast cells did not differ at the foreign body site in comparison to an alternate site. Intracellular bacteria were found significantly more often in lavage samples at the site of the foreign body (17/21, 81%) than at the alternate site (7/17, 41%) (P 5.018). All samples that had bacteria detected cytologically at the alternate site also had bacteria identified in the lavage sample from the foreign body site. BAL was performed in only 3 cats. Cytologic assessment was available for 2 cats and indicated 500 cells/ml for each with 11 and 24% neutrophils and intracellular bacteria visualized. BAL fluid was submitted for aerobic culture in 24 dogs, and positive results were obtained in 18 cases (75%). Three of 6 dogs that lacked bacterial growth from aerobic culture and did not display intracellular bacteria were on antibiotics at the time of sampling; however, 4 other dogs that were on antibiotics at the time of sampling had positive bacterial cultures and intracellular bacteria visualized. Aerobes isolated more than once included Pasteurella (10/18), Streptococcus (5/18), Staphylococcus (4/18), Actinomyces (4/18), E. coli (4/18), and Corynebacterium (2/18). Mycoplasma culture was submitted in 23/25 cases and was positive in 3 (13%). In 2 of 3 samples, Mycoplasma was the only organism isolated, however 1 isolate was from a dog that had been on amoxicillin-clavulanate and potentiated sulfonamide up until 2 days before bronchoscopy. BAL fluid from 23/25 samples was submitted for anaerobic culture, and bacterial growth occurred in 11/23 (48%) cases, 4 of which animals were on antibiotics at the time of sampling. Anaerobic bacteria that could be speciated included Bacteroides/Prevotella (6/11), Fusobacteriuam (3/11), and Peptostreptococcus (4/11). Cultures in 3 cats yielded no growth in 1 and both aerobes (enterics, Pasteurella) and anaerobes (Fusobacterium, Anaerobius) in the remaining 2 samples. No Mycoplasma spp. were isolated from cats with airway foreign bodies. Discussion Flexible bronchoscopy was successful in removing airway foreign bodies in dogs (24/28, 86%) and was

5 Bronchoscopy for Foreign Body Removal 1067 Fig 4. Bronchoalveolar lavage fluid (100) obtained from a dog at the site of foreign body removal (A) and at an alternate site (B). Differential cell percentages were significantly different at the 2 sites. reasonably effective in cats (2/5, 40%) in this study, with an overall success rate of 79%. This success rate compares favorably with bronchoscopic management of foreign body aspiration in infants and children, as well as in previous studies in dogs, where success rates over 90% were reported. 2,6 8 Success rates in adults are similar, with 75% success in 1 report utilizing both flexible and rigid bronchoscopes. 9 In human medicine, when flexible bronchoscopy is not successful, rigid bronchoscopy is often employed 7 or postbronchoscopy CT may be performed to identify the foreign body. 10 Surgery was chosen as the initial intervention in 4 dogs, and this decision typically was based on the presence of pneumothorax or lobar consolidation. Surgical therapy rarely is pursued in human medicine even when life-threatening complications of pneumothorax or pulmonary abscessation are identified. 11 It is impossible to predict whether repeat bronchoscopy after advanced imaging might have been successful in some of the cases examined here that required surgery for resolution of disease, but this approach might be considered in selected circumstances. In this study, a slight preponderance of male animals was affected (59%) and this finding has also been reported in affected infants and adult humans. 8,11,12 Also, foreign bodies were found most commonly in the right bronchus in dogs, similar to reports in both infants and adult humans. 6,8 This observation is most likely related to a straighter anatomical alignment of the right mainstem bronchus with the trachea in comparison to the left mainstem bronchus. In cats, the most common location was the trachea or carina, and this finding was expected given the much narrower luminal diameter of feline as compared with canine bronchi even when considering animals of similar body weight. In animals examined here, foreign material was most often organic, with foxtail awns, small branches, and plant clippings reported most commonly. In infants, seeds and peanuts are reported most commonly whereas in adults, plastic material and pins are retrieved most commonly. 6,11,12 One limitation of this study is that we could not confirm that the foreign body was solely responsible for clinical signs of cough or respiratory distress; however, signs resolved after removal of the foreign body in each case. Interestingly, the duration of clinical signs did not impact the success of bronchoscopy in foreign body retrieval. Similar success has been reported infants with an onset of signs 1 day to 3 months before presentation, 12 although 1 study suggested a higher complication rate in infants presented to the hospital more than 2 days after a witnessed aspiration event. 13 Duration of clinical signs in adults with foreign body aspiration is less comparable to our patient population because of the adult s ability to recall and report an aspiration event within 24 hours in most cases. 8 Radiographic findings in foreign body aspiration have been reported as focal pulmonary to interstitial opacities, pneumothorax, pleural effusion, or pleural thickening. This same study identified a strong correlation between localization of foreign bodies using multimodality imaging studies and anatomic localization. 1 In the current study, which evaluated only radiographs, typical findings were present in only 2/3 of cases evaluated, whereas 1/3 of radiographs showed diffuse infiltrates or were considered normal. Radiographs with typical radiographic features resulted in a high rate of success (86%) in diagnosing the correct location of the foreign body. In the cases where probability was considered low for the presence of a foreign body, the majority of radiographic evaluations of location for the foreign body were incorrect due to lack of localizing features. In adults, radiographic location correlated with anatomic location in 68% of cases in 1 study, but similar to our findings, radiographs were normal in 28% of adult cases 8 and in 11 28% of infant cases. 12,13 This observation indicates that whereas radiographs are helpful in ruling in a foreign body and can localize the foreign body, the absence of specific radiographic findings does not rule out a foreign body. BAL fluid cytology assessment indicated greater inflammation and more frequent presence of intracellular bacteria at the site of lavage than at an alternate site. This finding is not unexpected because one of the advantages of bronchoscopic BAL is collection of a directed airway sample. However, collection of a more global sample such as that obtained during a tracheal wash may underestimate the degree of inflammation associated with the presence of a foreign body or might fail to detect infection (intracellular bacteria). Additional studies would be required to test this hypothesis. Bacteria were visualized less often in BAL fluid from a region distant to the

6 1068 Tenwolde et al foreign body compared with fluid from direct lavage of the foreign body site, and performing cytology on an additional lavage sample in cases with obvious foreign material within the airway cannot be definitively advised. Samples submitted for cultures were pooled from all sites sampled, thus site differences in microbiologic isolation could not be addressed in this study. As expected, multiple bacterial species were isolated from dogs with foreign bodies, with aerobes isolated from 72%, anaerobes from 48% and Mycoplasma spp. from 13%. Mixed bacterial species were identified in over half of the dogs that had cultures performed for specific organisms, and bacteria could be isolated even in dogs on antibiotic therapy. This finding supports the clinical importance of performing aerobic, anaerobic, and Mycoplasma cultures in animals with foreign body pneumonia, regardless of whether or not antibiotics currently are being used. Another limitation of this study is the relatively small sample size of cats with airway foreign bodies. This likely reflects environmental exposure, behavioral differences, and outdoor activity level differences between dogs and cats. Given the small number of cats in this study, it is difficult to compare the success of bronchoscopy in cats versus dogs, although it was possible to avoid surgery in 2/5 cats examined here. In conclusion, flexible bronchoscopy was a minimally invasive and successful therapeutic option for the safe and effective removal of bronchial foreign bodies in dogs and cats in this study. This is likely a reflection of the wide range of equipment available at our referral hospital as well as the medical, technical, and anesthetic expertise at the hospital. Our high success rate might also be related to the type of foreign material most commonly encountered in our geographic region (ie, primarily linear foreign bodies such as grass awns and sticks). Large, irregularly shaped, or obstructive material such as teeth, toys, or endotracheal tubes were not discovered in cases examined here, and these can be more challenging to remove with bronchoscopic equipment. We conclude that bronchoscopy should be utilized as a primary therapy option unless clear indication for surgical intervention is identified by imaging studies. Bronchoscopy may be more successful in dogs than cats because of small airway diameter limitations in cats. In this study, and similar to other studies, thoracic radiographs are helpful in most cases to support the diagnosis of a foreign body, however they must be interpreted with caution because normal or nonlocalizing radiographs do not rule out the presence of a foreign object in the airway. This study determined that the degree and nature of localized inflammation and infection at the site of the foreign material differed from alternate sites in the lungs, indicating that BAL should be directed specifically at the site of foreign body removal in order to best identify infection and guide antimicrobial therapy after foreign body removal. Footnotes a 2.5 mm70 cm Karl Storz VB fiberoptic endoscope, Goleta, CA; Olympus BF3C mm55 cm videoendoscope, Melville, NY b Olympus P20D 5.0 mm55 cm fiberoptic endoscope, Melville, NY; Pentax FG16X or V 5.5 mm85 cm fiberoptic endoscope, Montvale, NJ; Olympus XP m fiberoptic endoscope, Melville, NY References 1. Schultz RM, Zwingenberger A. Radiographic, computed tomographic, and ultrasonographic findings with migrating intrathoracic grass awns in dogs and cats. Vet Rad Ultrasound 2008;49: Lotti U, Niebauer GW. Trachebronchial foreign bodies of plant origin in 153 hunting dogs. J Am Anim Hosp Assoc 1992;14: Tivers MS, Hotston Moore A. Tracheal foreign bodies in the cat and the use of fluoroscopy for removal: 12 cases. J Small Anim Pract 2006;47: Hawkins EC, DeNicola DB, Kuehn NF. Bronchoalveolar lavage in the evaluation of pulmonary disease in the dog and cat. State of the art. J Vet Intern Med 1990;4: Padrid PA, Feldman BF, Funk K, et al. Cytologic, microbiologic, and biochemical analysis of bronchoalveolar lavage fluid obtained from 24 healthy cats. Am J Vet Res 1991;52: Saki N, Nikakhlagh S, Rahim F, Abshirini H. Foreign body aspirations in infancy: A 20-year experience. Int J Med Sci 2009;6: Tang LF, Xu YC, Wang YS, et al. Airway foreign body removal by flexible bronchoscopy: Experience with 1027 children during World J Pediatr 2009;5: Yadav SP, Singh J, Aggarwal N, Goel A. Airway foreign bodies in children: Experience of 132 cases. Singapore Med J 2007;48: Ramos MB, Fernández-Villar A, Rivo JE, et al. Extraction of airway foreign bodies in adults: Experience from Interact Cardiovasc Thorac Surg 2009;9: Shin SM, Kim WS, Cheon JE, et al. CT in children with suspected residual foreign body in airway after bronchoscopy. Am J Roentgenol 2009;192: Li Y, Wu W, Yang X, Li J. Treatment of 38 cases of foreign body aspiration in children causing life-threatening complications. Int J Pediatr Otorhinolaryngol 2009;73: Shubha AM, Das K. Tracheobronchial foreign bodies in infants. Int J Pediatr Otorhinolaryngol 2009;73: Shlizerman L, Mazzawi S, Rakover Y, Ashkenazi D. Foreign body aspiration in children: The effects of delayed diagnosis. Am J Otolaryngol 2009 April 22 (Epub ahead of print).