Bronchoscopy and BAL are minimally invasive techniques used in human and veterinary medicine to visually evaluate the trachea and bronchi and to sample the bronchoalveolar spaces, respectively.1–4 Common indications for BAL include chronic cough, unexplained abnormalities on thoracic radiographs, and pulmonary neoplasia.3,4 After wedging the bronchoscope in a bronchus, BAL is performed by infusion of sterile saline (0.9% NaCl) solution into the small bronchi and alveoli followed by the immediate retrieval of the infusate (BALF) for analysis.2–4 Unfortunately, certain technical aspects of the BAL procedure vary, and there is no standardized protocol for BAL in dogs.3–7 These technical aspects, such as aliquot volume used and the negative pressure applied for aspiration, affect the quality of BALF from healthy dogs.6,7 To our knowledge, the method of BALF retrieval from dogs has not been evaluated. It has been hypothesized that SPA may result in decreased cell viability in BALF because excessive negative pressure can cause cell lysis.2 Results of another study8 performed by our research group indicate that BALF from healthy dogs obtained by automated SPA has a better diagnostic quality than that obtained by MA through polyethylene tubing. The purpose of the study reported here was to compare quantity and quality of BALF samples obtained by MA with a handheld syringe with those of BALF samples obtained by automated SPA in healthy dogs. We hypothesized that the method of aspiration would affect BALF sample quality and potentially influence the diagnostic yield of BALF in dogs.
Materials and Methods
Animals—Dogs included in the study were required to be free of respiratory tract disease. Thirteen adult Beagles were determined to be healthy on the basis of medical history and results of physical examination, CBC, serum biochemical analysis, and evaluation of lateral and ventrodorsal orthogonal thoracic radiographs by a board-certified veterinary radiologist (HD). The dogs regularly received prophylactic anthelminthic treatment but had not received any other medications within 30 days prior to study initiation. Dogs were cared for at a Canadian Council on Animal Care–accredited facility in accordance with the council's guidelines, and the study protocol was approved by the University of Guelph Animal Care Committee.
Anesthesia—Each dog was sedated, and anesthesia was induced and maintained as described for a previous study.8 Intravenous fluid therapy was administered, and supplemental oxygen was provided through a soft sterile cathetera that was placed in the trachea via the oral cavity. During anesthesia, each dog was instrumented with a continuous ECG and pulse oximeter, and the heart rate and rhythm, respiratory rate, percentage oxygen saturation, and body temperature were monitored. Clinically normal oxygen saturation was defined as ≥ 95%. The experimental procedure was aborted and a dog was allowed to recover from anesthesia if the oxygen saturation decreased to < 90% for 10 minutes or < 85% for 5 minutes or bradycardia (< 60 beats/min) that did not respond to standard interventions developed.
BAL—Each dog was positioned in sternal recumbency during the experimental procedure. Bronchoscopy was performed in a routine manner9 and recorded with a video bronchoscope.b The trachea and all lung lobes, including second- and third-generation bronchi, were visually examined following infusion of 2 mL of sterile 0.2% lidocaine solution at the carina to abate coughing during the examination.
The SPA and MA BAL techniques were performed in the caudal lung lobes in each dog; the aspiration technique used in each caudal lung lobe and the order in which the techniques were performed were randomly allocated by use of a random number table. The caudal lung lobes were lavaged with a weight-based volume (2 mL/kg, divided into 2 aliquots), with the second aliquot infused immediately after retrieval of the first aliquot.7 To perform SPA, the tip of the bronchoscope was wedged into a second-generation bronchus and warmed (37°C) sterile saline solution was infused through the biopsy channel followed by injection of 4 mL of air to empty the channel. Immediately after infusion, pulsatile aspiration with a maximum negative pressure of 5 kPa (37.5 mm Hg) was performed with a portable automated surgical suction pumpc connected directly to the suction valve of the bronchoscope by a disposable suction trap.d The MA technique was performed on the contralateral caudal lung lobe by wedging the tip of the bronchoscope into a second-generation bronchus, and the infusion of warmed sterile saline solution and air were performed as described for the SPA technique. A 35-mL syringe was directly attached to the bronchoscope's biopsy channel, and gentle pulsatile MA was applied, with the syringe emptied of air as necessary to continue aspiration. For both techniques, aspiration was continued until fluid could no longer be recovered. Between use in dogs, the bronchoscope was cleaned and sterilized in accordance with a standard cold sterilization method.e–g
The duration of BAL (time from infusion of saline solution until the end of aspiration for the second aliquot), lowest percentage oxygen saturation, and amount of BALF retrieved were recorded for each technique. A minimum retrieval of 33% of the BAL infusate volume was defined as a diagnostic BALF sample.10 Dwell time (time from infusion of saline solution to the first attempt at aspiration) was < 20 seconds for each collection.
Examination of BALF—Immediately after collection, macroscopic surfactant was indirectly assessed by visual estimation of foam in each sample and assignment of a surfactant score on a dichotomized scale of absent (0) or present (1). Each BALF sample was then identified by a unique code of 3 letters, placed on ice, and processed for analysis within 120 minutes after collection. Samples were not filtered during processing. Total nucleated cell counts were determined with an automated cell counterh by use of electrical impedance; the typical count precision for this counter was 2% to 3% counting variability, with a lower limit for particle detection of 1 μm. A 200-μL aliquot of each sample was cytocentrifugedi for 6 minutes at 180 × g, and additional direct smears were prepared from BALF that was centrifuged for 5 minutes at 500 × g. Slides were stained with Wright stain, and differential cell counts of 400 leukocytes were performed on cytocentrifuge preparations examined with a microscopej set at 400× magnification by a board-certified veterinary clinical pathologist (DB) who was unaware of the technique used to obtain each sample. For each slide, the pathologist also assessed 5 semiquantitative variables associated with sample and slide quality as described,8 and scores ≥ 2 for cellularity and cellular preservation were required for a BALF sample to be considered of adequate diagnostic quality. Additionally, the cells on each slide were evaluated for evidence of erythrophagocytosis; scores > 2 for epithelial cells or RBCs (without erythrophagocytosis) were considered indicative of excessive suction or trauma and a low-quality BALF sample. The phospholipid concentration in the BALF samples was not determined because an insufficient amount of each sample to perform the method described by Bligh and Dyer11 remained after the other analyses were completed.
Statistical analysis—Data distributions were assessed for normality with a Shapiro-Wilk test. For continuous outcomes that were normally distributed (duration of BAL, percentage BALF retrieval, percentage oxygen saturation, percentage lymphocytes, and percentage macrophages), comparisons between MA and SPA were performed with a paired student t test. For continuous outcomes that were not normally distributed (TNCC, percentage neutrophils, and percentage eosinophils) and ordinal outcomes (scores for epithelial cells, cellularity, and cellular preservation), comparisons between MA and SPA were performed with a Wilcoxon signed rank test. For analysis purposes, results were dichotomized as absent (0) or present (1) for macroscopic surfactant, RBCs, bacteria, and mast cells, and comparisons between MA and SPA were performed with a McNemar test. Statistical analyses were performed with a statistical software program,k and values of P < 0.05 were considered significant.
Results
Animals—Thirteen adult Beagles were evaluated for inclusion in the study. During physical examination, 1 dog had bilateral purulent nasal discharge and was excluded from the study. The 12 dogs included in the study consisted of 3 sexually intact males, 5 castrated males, and 4 spayed females with a mean ± SD age of 9.7 ± 0.8 years (range, 8.1 to 10.8 years) and weight of 10.8 ± 2.1 kg (range, 8.1 to 15.5 kg). For all study dogs, body temperature and heart and respiratory rates were within reference limits during physical examination and results of CBCs and serum biochemical analyses were unremarkable. Three dogs had a left-sided systolic murmur (1 grade II/VI, 1 grade III/VI, and 1 grade IV/VI) auscultated during physical examination and evidence of mild to moderate left atrial enlargement on thoracic radiographs. Because none of those dogs had clinical signs of heart disease or evidence of pulmonary edema on thoracic radiographs, it was suspected they had compensated left mitral valve disease, and they were not excluded from the study.
Bronchoscopy and BAL—Bronchoscopic examination findings were within reference limits for 8 of 12 dogs. In 2 dogs, the bronchial mucosa had a mild nodular appearance in the left lung lobes but not in the right lung lobes. One dog had a diffuse, mildly increased amount of bronchial mucous, compared with that of the other study dogs, and airway collapse was observed during BAL with both MA and SPA in 1 dog. The duration of the BAL did not differ significantly between MA and SPA (Table 1) and ranged from 46 to 107 seconds for MA and 43 to 137 seconds for SPA.
Mean or median values for variables evaluated following bronchoscopic BAL of 12 healthy adult Beagles in which BALF samples were obtained by MA and SPA.
Technique | |||||
---|---|---|---|---|---|
Variable | MA | SPA | SD or IQR | Mean or median distribution of differences* | P value |
Duration of BAL (s) | 78.4 | 65.9 | 21.5 | −12.5 | 0.069 |
Lowest oxygen saturation during BAL (%) | 93 | 92 | 0.04 | −1.0 | 0.52 |
Macroscopic surfactant score† | 1 | 1 | — | — | 1.0 |
Percentage of BALF retrieved (%) | 46.5 | 69.5 | 8.8 | 23.0 | < 0.001 |
BALF quality score Cellularity‡ | 3.5 | 4 | 1.0 | 0 | 0.12 |
Cellular preservation§ | 4 | 4 | 0.5 | 0 | 0.62 |
Epithelial cells‖ | 1 | 0 | 0.5 | 1 | 0.91 |
RBCs† | 0 | 0 | — | — | 1.00 |
Bacteria† | 0 | 0 | — | — | 0.48 |
TNCC (cells/μL)# | 200 | 200 | 50 | 100 | 0.02 |
BALF differential cell count# Macrophage (%) | 72 | 75 | 8 | 3 | 0.24 |
Neutrophil (%) | 2 | 2 | 2 | −1 | 0.37 |
Lymphocyte (%) | 24 | 19 | 9 | −5 | 0.12 |
Eosinophil (%) | 2 | 0 | 6 | 1 | 0.56 |
Mast cell† | 0 | 0 | — | — | 1.00 |
A positive value for the mean or median distribution of differences indicates that the value for samples obtained by SPA was higher than the value for samples obtained by MA, whereas a negative value for the mean or median distribution of differences indicates that the value for samples obtained by SPA was lower than the value for samples obtained by MA.
Results dichotomized as absent (0) or present (1).
Scored on the basis of cytologic evaluation on a scale of 0 to 4 (0, < 10 cells/slide; 1, 10 to 100 cells/slide; 2, 101 to 200 cells/slide; 3, 201 to 500 cells/slide; and 4, > 500 cells/slide).
Scored on the basis of cytologic evaluation on a scale of 0 to 4 (0, < 10% well-preserved cells/slide; 1, 10% to 25% well preserved cells/slide; 2, 26% to 50% well-preserved cells/slide; 3, 51% to 80% well-preserved cells/slide; and 4, > 80% well-preserved cells/slide).
Scored on the basis of cytologic evaluation on a scale of 0 to 4 (0, absent; 1, < 50 cells/slide; 2, 51 to 100 cells/slide; 3, 101 to 200 cells/slide; and 4, > 200 cells/slide).
Represents data from only 9 dogs; data for 2 dogs were excluded because they had asymmetric bronchoscopic abnormalities, and data for another dog were excluded because it had evidence of septic suppurative inflammation on cytologic evaluation.
— = Not determined.
Both MA and SPA were performed during the same bronchoscopy examination in all dogs; the aspiration technique used in each caudal lung lobe and the order in which the techniques were performed were randomly allocated.
Adverse effects of BAL—Following BAL, mild bilateral bruising of the bronchial mucosa was observed in 1 dog. Transient (< 2-minute) decreases in the percentage oxygen saturation during MA and SPA were observed in 8 and 6 dogs, respectively; however, none of the dogs required intervention. The mean lowest percentage oxygen saturation did not differ significantly between MA (93%) and SPA (92%).
BALF analysis—Surfactant was macroscopically present in 12 of 12 samples obtained with SPA and 11 of 12 samples obtained with MA, and the surfactant scores did not differ significantly between MA and SPA. The mean ± SD percentage of BALF retrieved by SPA (69.5 ± 6.5%) was significantly (P < 0.001) greater than that retrieved by MA (46.5 ± 10.2%), with a mean ± SD difference of 23.0 ± 12.6%. For 3 BALF samples (2 obtained by MA and 1 obtained by SPA), no TNCCs were detected by electrical impedance; there was no association between TNCC and visual assessment of surfactant. In fact, the cytocentrifuge preparations of those 3 samples had adequate cellularity for cytologic evaluation such that TNCCs were manually estimated by the clinical pathologist (DB) and retained in the analysis. However, BALF samples from 3 dogs were excluded from the TNCC and differential cell count analyses. Two dogs had asymmetric bronchoscopic abnormalities and 1 dog had respiratory tract disease as evidenced by a markedly increased TNCC from reference limits and septic suppurative inflammation identified during cytologic evaluation of one of its BALF samples.
For all study dogs, the median cellularity score for samples obtained by MA (3.5; range, 2 to 4) did not differ from that for samples obtained by SPA (4; range, 3 to 4). All cellular preservation scores were 3 or 4, and the median cellular preservation score did not differ significantly between MA and SPA. Similarly, the scores for RBCs, bacteria, and epithelial cells did not differ significantly between MA and SPA.
For the 9 study dogs for which TNCC and differential cell counts were analyzed, the median TNCC was 200 cells/μL (range, 100 to 300 cells/μL; IQR, 100 cells/μL) in BALF obtained by MA and 200 cells/μL (range, 200 to 400 cells/μL; IQR, 100 cells/μL) in BALF obtained by SPA, with a median distribution of differences in TNCC between SPA and MA of 100 cells/μL (IQR, 50 cells/μL), which indicated that the TNCC in BALF samples obtained by SPA was significantly (P = 0.02) greater than that in BALF samples obtained by MA. Conversely, the respective differential counts of macrophages, neutrophils, lymphocytes, eosinophils, and mast cells did not differ between BALF samples obtained by MA and SPA.
Discussion
Results of the present study indicated that BAL with SPA recovered a greater percentage of BALF that had a higher TNCC than did MA. Collection of BALF by MA is the most commonly used technique in veterinary medicine3–5,7; however, results of another study8 conducted by our research group suggest that retrieval of BALF by SPA yielded samples of higher quality, compared with BALF samples obtained by MA through polyethylene tubing. The present study differed from that study8 in that the MA did not include the use of polyethylene tubing. The surfactant score8 was refined in consideration of the fact that it is currently unknown whether the method of fluid retrieval influences the amount or characteristics of foam and therefore the volume of bubbles in BALF. It should also be emphasized that no evidence-based recommendations currently exist regarding acceptable limits for cellular preservation or RBCs, and these scores8 were created on the basis of expert opinion. Yet, the results of both the present study and the previous study8 suggest that SPA provides BALF samples of higher diagnostic quality than does MA.
In the present study, SPA resulted in retrieval of a significantly greater percentage of BALF than did MA. The higher the percentage of BALF retrieved, the greater the likelihood that the BAL successfully sampled the alveoli as well as the bronchi.12 In human patients, the use of small-volume aliquots with a small percentage of fluid retrieval is considered a bronchial wash instead of an alveolar wash.12,13 Recovery of a minimum of 33% to 40% of the BALF is considered necessary for diagnostic purposes in human patients and dogs2,10; therefore, we considered retrieval of < 33% of BALF insufficient for diagnostic purposes. In the present study, < 33% of the BALF was retrieved by MA in 1 of 12 dogs, whereas > 59% of the BALF was retrieved by SPA in all 12 study dogs. Consequently, the mean percentage of BALF retrieved was significantly greater for SPA, compared with that for MA.
In healthy dogs, decreased BALF retrieval could be caused by improper placement of the bronchoscope (eg, abuttal of the bronchoscope tip against a bronchial wall or improper wedging of the bronchoscope tip).2,14 In the present study, placement of the bronchoscope within the bronchus was visually assessed to ensure that the tip was not occluded against the bronchial wall prior to all BALs. Thus, bronchoscope occlusion was an unlikely cause of the low rate of BALF retrieval by MA. Failure to achieve a seal between the bronchoscope tip and the bronchial wall will result in the retrieval of air rather than infusate during the aspiration phase of a BAL.14 During aspiration, this seal can be lost by movement of the bronchoscope. Movement of the bronchoscope is more likely to occur during MA than it is during SPA because MA is performed by an assistant, whereas SPA is performed by the clinician who is operating the bronchoscope. Moreover, it is possible that SPA results in a better seal between the wedged bronchoscope and bronchial wall because of the creation of greater negative pressure than MA creates.
Occasionally, BALF cannot be retrieved despite adequate wedging of the bronchoscope into the bronchus, and in those instances, it is recommended that the bronchoscope be retracted a few millimeters.2,14 In the present study, this was done prior to discontinuation of BAL for both aspiration techniques in all dogs; therefore, it is unlikely that this was the cause of a nondiagnostic amount of BALF being obtained during MA in 1 dog. Furthermore, it is also unlikely that use of the handheld 35-mL syringe was insufficient to create the negative pressure needed to retrieve BALF because this technique has been successfully used in another study.3 Assuming that the amount of BALF retrieved was independent of wedging the bronchoscope into the bronchus, it is likely that the high percentage of BALF retrieved by SPA was the result of the creation of a more consistent negative pressure without concurrent airway collapse, compared with that created by MA.
In the present study, 3 BALF samples had TNCCs of 0 as determined by electrical impedance. However, the cytocentrifuge preparations for those 3 samples contained a sufficient number of cells for differential counts to be performed. Cell lysis as a cause for the 3 acellular BALF samples by electrical impedance was considered unlikely because, if cell lysis were the result of BAL technique or BALF processing, inadequate cell numbers would have been expected on cytocentrifuge preparations. In fact, the opposite was identified in the present study. This suggested that laboratory error might have caused the failure of the automated cell counter to accurately count the TNCCs in those samples. Mucus in BALF can cause cells to clump and occlude the aperture of the automated cell counter, resulting in artificially decreased cell counts. Regardless, one of the apparently acellular BALF samples acquired by MA had a low cellularity score (2) on cytocentrifuge preparation and would have been classified as nondiagnostic on the basis of a low percentage of BALF retrieval (25%).
Although the BALF obtained by SPA had significantly higher median distribution of TNCC than did that obtained by MA in the present study, the cellularity scores, including the cellular preservation scores, did not differ significantly between BALF samples obtained by SPA and MA. This constrasts with concerns expressed by other clinicians2 that the increased negative pressure created by SPA, compared with that created by MA, would cause cellular damage. Additionally, the differential cell counts did not differ significantly between samples obtained by SPA and MA, which suggested that no particular type of leukocyte was preferentially destroyed during aspiration by SPA or MA, despite the fact that the disposable suction trap (polystyrene) and handheld syringe (polypropylene) were made of different materials. Similarly, the RBC, epithelial cell, and bacterial scores did not differ significantly between samples obtained by SPA and MA, which suggested that one technique was not any more traumatic to the bronchial mucosa than was the other. Thus, in regard to the quality and preservation of cells, BALF samples obtained by SPA were comparable to BALF samples obtained by MA.
The question remains as to whether the significantly increased percentage of BALF retrieved and TNCC in samples obtained by SPA, compared with those in samples obtained by MA, are clinically relevant for dogs with respiratory tract disease. An increased number of TNCC is unlikely to increase the likelihood of identification of neoplastic cells or pathogenic organisms in BALF because these changes are classified as either present or absent during cytologic analysis.15 In human patients, studies16,17 that assessed the association between diagnostic yield and percentage of BALF retrieved have yielded conflicting results. Results of 1 study,16 in which BALF samples obtained from human patients by MA with and without tubing were compared, indicate that diagnostic yield was highest for BALF samples obtained by MA via tubing, and diagnostic yield was positively correlated with percentage of BALF retrieved. Conversely, results of another study,17 in which BALF samples obtained from human patients with hematologic or nonhematologic malignant cancer were compared, indicate that the amount of BALF retrieved was not correlated with diagnostic yield. However, investigators of that study17 only assessed BALF for evidence of infectious pneumonia or malignancy, and this might have biased their results. For dogs with respiratory tract disease, further evaluation of BAL aspiration techniques is necessary to assess whether aspiration technique affects the diagnostic yield of BAL.
Strengths of the present study included randomly allocating the caudal lung lobe and order in which SPA and MA were performed, having the same operators (KSW and AMND) perform all BALs, and ensuring that the pathologist who performed the cytologic evaluations was unaware of the aspiration technique used to obtain each sample. Another strength of this study was the comparison BALF samples obtained by MA through a handheld syringe attached directly to the bronchoscope biopsy channel with those obtained by SPA because MA is the aspiration technique most frequently used to obtain BALF from dogs.3–5,7
Limitations of the present study included the small study population that consisted of healthy dogs of similar age, body weight, and breed and the fact that surfactant was not directly measured in BALF samples. These limitations might restrict the extrapolation of results to dogs with respiratory tract disease or different signalment. For example, wedging a bronchoscope in a bronchus is more difficult in larger dogs2,9,14; therefore, some clinicians will use an endoscope with a larger diameter than that used in the present study for BAL in large-breed dogs. Different working channel diameters and lengths (eg, longer endoscopes) will likely alter negative pressure achieved during the aspiration phase of BAL.
For the healthy dogs of the present study, SPA resulted in retrieval of a significantly greater percentage of BALF that had a greater TNCC than did MA. However, cytologic quality did not differ significantly between BALF samples obtained by SPA and MA. These results suggested that SPA might improve the diagnostic yield of BAL in dogs, but further evaluation of BAL aspiration technique on BALF quantity and quality in dogs with respiratory tract disease is warranted.
ABBREVIATIONS
BAL | Bronchoalveolar lavage |
BALF | Bronchoalveolar lavage fluid |
IQR | Interquartile range |
MA | Manual aspiration |
SPA | Suction pump aspiration |
TNCC | Total nucleated cell count |
Arnolds dog catheter with female Luer mount, 8F × 50 cm, Smith Medical International, Hythe, Kent, England.
Olympus BF type 1T160 video bronchoscope, outer diameter 6.0 mm, working length 60 cm, biopsy channel inner diameter 2.8 mm, Olympus Canada Inc, Richmond Hill, ON, Canada.
Olympus SSU-2 endoscopic aspiration pump, Olympus Canada Inc, Richmond Hill, ON, Canada.
Kendall Luki 20-mL (6.25-inch) disposable aspirating tube, Tyco Healthcare Group, Mansfield, Mass.
Video bronchoscope cleaning and cold sterilization protocol, Olympus Canada Inc, Richmond Hill, ON, Canada.
Endozime, dual enzymatic cleaning, Ruhof Corp, Mineola, NY.
Glutacide, Pharmax Ltd, Etobicoke, ON, Canada.
Z2 Coulter counter, Beckman Coulter Inc, Mississauga, ON, Canada.
Shandon Cytospin 4, Thermo Fisher Scientific Inc, Waltham, Mass.
Olympus BX53 system microscope, Olympus Canada Inc, Richmond Hill, ON, Canada.
SAS, version 9.1, SAS Institute Inc, Cary, NC.
References
1. Klech H, Pohl W. Technical recommendations and guidelines for bronchoalveolar lavage (BAL). Eur Respir J 1989; 2:561–585.
2. Hawkins EC. Bronchoalveolar lavage. In: King LG, ed. Textbook of respiratory disease in dogs and cats. St Louis: Saunders Elsevier, 2004;118–128.
3. 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; 4:267–274.
4. Hawkins EC, DeNicola DB, Plier ML. Cytological analysis of bronchoalveolar lavage fluid in the diagnosis of spontaneous respiratory tract disease in dogs: a retrospective study. J Vet Intern Med 1995; 9:386–392.
5. McCauley M, Atwell RB, Sutton RH, et al. Unguided bronchoalveolar lavage techniques and residual effects in dogs. Aus Vet J 1998; 76:161–165.
6. Mordelet-Dambrine MM, Arnoux AA, Stanislas-Leguern GG, et al. Processing of lung lavage fluid causes variability in bronchoalveolar cell count. Am Rev Respir Dis 1984; 130:305–306.
7. Melamies MA, Jarvinen AK, Seppala KM, et al. Comparison of results for weight-adjusted and fixed-amount bronchoalveolar lavage techniques in healthy Beagles. Am J Vet Res 2011; 72:694–698.
8. Woods KS, Defarges AMN, Abrams-Ogg ACG, et al. Comparison between manual aspiration via polyethylene tubing and aspiration via a suction pump with a suction trap connection for performing bronchoalveolar lavage in healthy dogs. Am J Vet Res 2013; 74:523–529.
9. Padrid PA. Laryngoscopy and tracheobronchoscopy of the dog and cat. In: Tams TR, Rawlings CA, eds. Small animal endoscopy. 3rd ed. St Louis: Elsevier Mosby, 2011;331–359.
10. Ward C, Walters EH. Bronchoalveolar lavage. In: Rogers DF, Donnelly LE, eds. Human airway inflammation: sampling techniques and analytical protocols. Totowa, NJ: Humana Press Inc, 2001;43–70.
11. Bligh EG, Dyer WJ. A rapid method of total lipid extraction and purification. Can J Biochem Physiol 1959; 37:911–917.
12. Lam S, Leriche JC, Kijek K, et al. Effect of bronchial lavage volume on cellular and protein recovery. Chest 1985; 88:856–859.
13. Baughman RP. Technical aspects of bronchoalveolar lavage: recommendations for a standard procedure. Semin Respir Crit Care Med 2007; 28:475–485.
14. Creevey KE. Airway evaluation and flexible endoscopic procedures in dogs and cats: laryngoscopy, transtracheal wash, tracheobronchoscopy, and bronchoalveolar lavage. Vet Clin North Am Small Anim Pract 2009; 39:869–880.
15. English K, Cowell RL, Tyler RD, et al. Transtracheal and bronchoalveolar washes. In: Cowell RL, Tyler RD, Meinkoth JH, et al, eds. Diagnostic cytology and hematology of the dog and cat. 3rd ed. St Louis: Mosby Elsevier, 2008;256–277.
16. Rosell A, Xaubet A, Agusti C, et al. A new BAL fluid instillation and aspiration technique: a multicenter randomized study. Respir Med 2006; 100:529–535.
17. Sampsonas F, Kontoyiannis DP, Dickey BF, et al. Performance of a standardized bronchoalveolar lavage protocol in a comprehensive cancer center: a prospective 2-year study. Cancer 2011; 117:3424–3433.