Presented in part at the European College of Veterinary Internal Medicine-Companion Animal, Seville, Spain, September 2011
Microbiologic and Cytologic Assessment of Bronchoalveolar Lavage Fluid from Dogs with Lower Respiratory Tract Infection: 105 Cases (2001–2011)
Article first published online: 30 JAN 2013
Copyright © 2013 by the American College of Veterinary Internal Medicine
Journal of Veterinary Internal Medicine
Volume 27, Issue 2, pages 259–267, March/April 2013
How to Cite
Johnson, L.R., Queen, E.V., Vernau, W., Sykes, J.E. and Byrne, B.A. (2013), Microbiologic and Cytologic Assessment of Bronchoalveolar Lavage Fluid from Dogs with Lower Respiratory Tract Infection: 105 Cases (2001–2011). Journal of Veterinary Internal Medicine, 27: 259–267. doi: 10.1111/jvim.12037
- Issue published online: 15 MAR 2013
- Article first published online: 30 JAN 2013
- Manuscript Accepted: 3 DEC 2012
- Manuscript Revised: 22 OCT 2012
- Manuscript Received: 18 JUN 2012
- Respiratory endoscopy;
- Thoracic radiology
Documentation of lower respiratory tract infection has relied on microbiologic and cytologic findings in airway fluid, but there is no gold standard for making a definitive diagnosis.
To report cytologic and microbiologic findings in dogs diagnosed with lower respiratory tract infection through evaluation by bronchoscopy and bronchoalveolar lavage.
A total of 105 dogs with spontaneous respiratory disease.
Retrospective case review of all dogs identified through the electronic medical record database that had bronchoscopy with bronchoalveolar lavage performed between 2001 and 2011. Results of bronchoalveolar lavage cytology and microbiology were evaluated in 510 dogs, and 105 cases with septic, suppurative inflammation or bacterial growth from cultures were examined further.
Bacteria were isolated from 89/105 aerobic cultures, 18/104 anaerobic cultures, and 30/99 Mycoplasma spp. cultures. The most common isolate was Mycoplasma spp. followed by Pasteurella sp., Bordetella sp, Enterobacteriaceae, and anaerobes. A single bacterial species was cultured from 44/99 dogs (44%) and multiple bacterial species were isolated from 55/99 dogs (56%). Suppurative inflammation with intracellular bacteria was identified cytologically in 78 of 105 dogs (74%). In 27 dogs that lacked cytologic evidence of sepsis, mixed (n = 18) and neutrophilic (n = 9) inflammation was reported, and Mycoplasma spp. (13/27) or Bordetella spp. (7/27) were most commonly isolated. Most aerobic bacteria were susceptible to routinely used antimicrobial drugs.
Conclusions and Clinical Importance
Confirmation of lower respiratory tract infection in dogs is challenging and organisms can be isolated from dogs in which bacteria are not detected on cytologic examination.
University of California, Davis Veterinary Medical Teaching Hospital
Lower respiratory tract infection is one of several differential diagnoses in dogs with cough, abnormal respiratory pattern or lung sounds, or radiographic infiltrates. Definitive diagnosis is assumed when microbial culture of an airway sample yields a positive result and there is supportive evidence of intracellular bacteria on cytology. One study suggested a threshold of quantitative bacterial growth exceeding 1.7 × 103 CFU/mL for a diagnosis of lower respiratory tract infection, and bacteria were observed cytologically in 71% of these cases. However, a diagnosis of respiratory infection is often made based on findings of airway culture alone in dogs with supportive clinical evidence of lower respiratory tract disease,[2-4] which could result in overdiagnosis of infection because the carina and large airways of dogs are not sterile.
The role of bronchoscopy for evaluation of lower respiratory disease in dogs has been well established, and bronchoalveolar lavage (BAL) fluid analysis allows differentiation among many respiratory diseases. In humans with ventilator-associated pneumonia, culture of BAL fluid more accurately identified infecting bacteria than culture of tracheal aspirates, resulting in more appropriate antimicrobial drug choice(s) and improved clinical outcome. Previous large-scale studies of microbes involved in lower respiratory tract infection in dogs have evaluated specimens collected by tracheal wash.[1, 2] Fluid characteristics of tracheal wash and BAL fluid are anticipated to differ because a tracheal wash collects fluid from the carina, while BAL samples the distal bronchoalveolar compartment. Infectious organisms, including bacteria and fungi, are more often found in cytology of BAL samples than in tracheal wash specimens.[6, 8]
The purpose of this study was to evaluate bacterial types involved in lower respiratory tract infection in dogs as determined by findings in bronchoalveolar lavage fluid, i.e., suppurative septic inflammation or positive bacterial cultures. In addition, we investigated susceptibility patterns of isolated bacteria. Information gained from this study will provide useful clinical guidance for management of dogs with suspected lower respiratory tract infection.
Materials and Methods
Microbiology results from bronchoscopy procedures performed in dogs at the William R. Pritchard Veterinary Medicine Teaching Hospital at the University of California, Davis (UCD-VMTH) between January 2001 and January 2011 were evaluated to identify cases with positive bacterial cultures. In addition, all dogs that had suppurative inflammation with intracellular bacteria (septic inflammation) in BAL fluid were scrutinized. Dogs with bacteria isolated only from enrichment broth or with rare (≤ 3) bacterial colonies on culture that also lacked observed organisms on cytologic assessment were excluded from further analysis. An additional exclusion criterion included the presence of oropharyngeal contaminants (Simonsiella bacteria or squamous cells). For the remaining cases, details were recorded for isolation of aerobic, anaerobic, and Mycoplasma spp. Susceptibility results for aerobic bacteria were tabulated.
Medical records were reviewed and clinical data were abstracted, including age, weight, duration of cough, temperature, and respiratory rate. Information on the type and duration of previous antimicrobial drug use was collected, and, specifically, use of antibiotics at the time of airway sampling was recorded.
Bronchoscopy and BAL
For animals with tracheal size large enough to accommodate a size 7 endotracheal tube, a 5–5.3 mm flexible bronchoscope1 was passed through a T adapter and gas anesthesia was employed. In smaller animals, anesthesia was performed with propofol (0.1–0.4 mg/kg/min, IV), oxygenation was maintained by jet ventilation at 180 breaths/min, and a 2.8–3.8 mm flexible bronchoscope2 was inserted directly through the trachea into the lungs.
Bronchoalveolar lavage was performed at one or more sites. After bronchoscopic examination of all airways, the endoscope was withdrawn from the respiratory tract, the exterior was wiped with sterile 0.9% saline-soaked gauze pads, and the biopsy channel was flushed with sterile saline to reduce airway contamination. The bronchoscope was returned to the site identified for lavage, with care taken not to contaminate the tip within the upper airways. An aliquot of warm, sterile saline was instilled through the biopsy channel of the endoscope and hand suction was applied to recover fluid that had contacted the bronchoalveolar space. Aliquot volume was determined by the endoscopist and ranged from 5 to 20 mL per site depending on body size.
BAL samples were analyzed by total and differential cell counts (based on a count of 200 cells) and cytologic assessment, including the presence of oropharyngeal contaminants, intracellular bacteria, and foreign material. Reference intervals used for canine BAL fluid were cell counts of 300–500 cells/μL comprised of less than 5–8% eosinophils, neutrophils, or lymphocytes and 65–85% macrophages along with the absence of intracellular bacteria. Samples with >8% neutrophils were classified as suppurative. When any number of neutrophils contained intracellular bacteria, the sample was categorized as septic.
BAL fluid was transported immediately to the Veterinary Microbiology Laboratory in the UCD VMTH and cultures were performed within 2 hours of collection using standard institutional technique. BAL fluid (2–3 drops of pooled sample from both sites) was plated onto 5% defibrinated sheep blood and MacConkey agars3 and incubated in the presence of 5% CO2 at 35°C for isolation of aerobic organisms. Prereduced anaerobic PRAS Brucella blood agar4 was used for anaerobic culture with incubation at 35°C under anaerobic conditions. Pleuropneumonia-like organism base with thallium acetate and penicillin G5 was used for Mycoplasma isolation with incubation at 35°C in 5% CO2. Bacterial growth was assessed in a semiquantitative fashion and reported as 1+, 2+, 3+, or 4+, designating the number of quadrants with bacterial growth. Standard biochemical methods were used to identify cultured bacteria. The number and species of isolates obtained from each dog were recorded. In some instances, multiple isolates of the same bacterial species, but with different morphologies or antibiograms, were isolated from a single case.
Bacterial susceptibility testing was performed according to standards established by the Clinical Laboratory Standards Institute using broth microdilution. Because of the time span of this study, susceptibility panels varied somewhat. Susceptibility interpretations of MIC testing were based on animal standards when available and when not available, interpretative criteria established for human medicine were used. Multidrug resistance was defined as resistance to 3 or more classes of antimicrobial drugs.
Clinical data were assessed for normality using the D'Agostino & Pearson omnibus test and are presented as mean ± standard deviation or median with range for nonparametric data. Cases were broadly separated into categories based on species of microbe isolated. Specifically, cases were grouped according to isolation of Mycoplasma, Bordetella, Enterobacteriaceae, Staphylococcus, Streptococcus, Pasteurella, Pseudomonas, or Stenotrophomonas, or anaerobes for comparison of clinical data including age, weight, duration of clinical signs, body temperature, and respiratory rate using a Kruskal–Wallis test6 for nonparametric data. The effect of age ≤1 year on Mycoplasma or Bordetella isolation was evaluated using Fisher's exact test. Significance was set at P < .05.
Bronchoscopy was performed on 510 dogs presented to the UC Davis Internal Medicine Service during the time period examined, and 266 dogs had no bacterial growth from BAL fluid in conjunction with cytologic findings of nonseptic inflammation. For the remaining 244 dogs, bacterial species were reported on culture or cytologic evaluation. Of these, 30 dogs were excluded because of bacterial growth in enrichment broth only and lack of intracellular bacteria on cytology. An additional 109 cases had rare bacterial colonies or evidence of oropharyngeal contamination in the absence of intracellular bacteria on cytology, which resulted in a total of 105 dogs with microbiologic or cytologic findings of lower respiratory tract infection. Aerobic culture results were available for review from all dogs. Mycoplasma culture was performed on samples from 99/105 and anaerobic culture from 104/105 dogs.
Median age of all dogs was 3 years (range 0.3–14.5 years) with 24/102 dogs (24%) less than 1 year of age (Table 1). Median body weight for all dogs was 15.7 kg (range 1–80 kg).
|All Dogs||Mycoplasma||B. bronchiseptica||Entero-bacteriaceae||Pasteurella spp||Staphylococcus||Streptococcus spp||P. aeruginosa||Anaerobes|
|Age (years)||3 (0.3–14.5)||1.8 (0.3–11)||1.0 (0.3–12)||9.0 (1–14.5)ab||4.2 (1–13)||6.0 (2–19.6)b||7.0 (2–12.5)b||8.5 (2–11)b||4.5 (1–14)ab|
|Number ≤1 year||24/102||12/30||17/23||1/14||3/15||0/5||0/13||0/7||3/18|
|Weight (kg)||15.7 (1–80)||6 (1–56)||6 (1.9–80)||22.7 (4.4–37.1)ab||30.2 (6.3–39.2)ab||26.6 (5.2–32.0)ab||27.7 (6.3–62.1)ab||26 (5.2–49.2)||23.2 (6.3–56.8)ab|
|Duration of cough (days)||120 (1–1800)||80 (2–1080)||60 (2–365)||90 (1–900)||180 (1–900)||242 (1–900)||120 (1–1000)||400 (60–1800)||75 (1–520)|
|Number with cough <7 days||20/105||4/27||1/22||6/16||2/15||1/5||2/11||0/6||4/18|
|Temperature (°F)||101.6 ± 1.3||101.7 ± 1.3||101.2 ± 1.1||101.4 ± 2.4||101.4 ± 0.7||102.1 ± 1.2||102.1 ± 1.4||101.5 ± 0.8||101.9 ± 1.3|
|Respiratory rate||41 ± 19||48 ± 36||38 ± 16||59 ± 30||35 ± 14||38 ± 3||43 ± 18||42 ± 3||37 ± 13|
Bacteria were isolated from 89/105 aerobic cultures, 18/104 anaerobic cultures, and 30/99 Mycoplasma cultures (Table 2). A single bacterial species was isolated from 44/99 (44%) dogs and multiple species were isolated from 55/99 (56%) dogs; no bacteria were isolated in 3 dogs. The most common isolates were Mycoplasma (30/99, 30%), B. bronchiseptica (23/105, 22%), Pasteurella (22/105, 21%), Enterobacteriaceae (21/105, 20%) and anaerobes (18/104, 17%). Organism identification was performed for only 1 of 30 Mycoplasma isolates and indicated M. cynos. In pure culture, the most commonly isolated organisms were B. bronchiseptica (10/105, 10%) and Mycoplasma (9/99, 9%). For the 13 dogs from which B. bronchiseptica was isolated in combination with other bacterial species, 9 dogs were coinfected with Mycoplasma and 4 dogs were coinfected with enteric organisms or Pasteurella.
|# of Specimens with Isolates (%)||# of Specimens Where Organism Was Isolated in Pure Culture|
|Enterobacteriaceae||21/105 (20%)||4/105 (4%)|
|E. coli||17/105 (17%)||4/105 (4%)|
|Pasteurella||22/105 (21%)||5/105 (5%)|
|P. canis||13/105||5/105 (5%)|
|Other Pasteurella spp.||5/105||0|
|Bordetella bronchiseptica||23/105 (22%)||10/105 (10%)|
|Mycoplasma spp.||30/99 (30%)||9/99 (9%)|
|Pseudomonas aeruginosa||6/105 (6%)||5/105 (5%)|
|Stenotrophomonas maltophila||2/105||2/105 (2%)|
|Actinomyces spp.||5/105 (5%)||0|
|Staphylococcus pseudointermedius/S. intermedius group||5/105 (5%)||3/105 (3%)|
|Streptococcus||13/105 (12%)||2/105 (2%)|
|S. viridans||4/105 (4%)||0|
|S. canis||4/105 (4%)||0|
|Unspeciated Streptococcus spp.||3/105 (3%)||0|
|S. zooepidemicus||2/105 (2%)||2/105 (2%)|
|Corynebacterium spp.||2/105 (2%)||0|
|Anaerobes||18/104 (17%)||4/104 (4%)|
Anaerobic bacteria were isolated from 18 dogs and were found in combination with aerobic bacteria in 13 dogs and with aerobes and Mycoplasma in 1 dog. Anaerobes that could be identified were Bacteroides/Prevotella (8/18), Peptostreptococcus anaerobius (5/18), Porphyromonas spp. (3/18), and 1 each of Propionobacterium spp., Clostridium spp., and Fusobacterium spp. In 4 dogs, only anaerobes were isolated (Prevotella bivi and Propionobacterium spp. in 1 dog each and mixed anaerobes in the remaining 2 dogs). In 11 of the 18 dogs from which anaerobes were isolated, an airway foreign body was identified. Anaerobic bacteria were not isolated from 10 additional dogs in which foreign bodies were identified.
Bronchoalveolar Lavage Cytology
Suppurative septic inflammation was identified cytologically in 78 of 105 dogs with lower respiratory tract infection. In 27 dogs that lacked cytologic evidence of intracellular bacteria, mixed inflammation (with neutrophils, lymphocytes, or eosinophils) was reported in 18/27 cases and suppurative inflammation was detected in the remaining 9 cases. In these 27 cases, bacterial culture yielded aerobes from 20 dogs (B. bronchiseptica in 7, Pasteurella in 4, Stenotrophomonas in 2, and miscellaneous bacteria in the remaining dogs). A pure culture of Mycoplasma was identified from 6 dogs in which intracellular bacteria were not observed and an anaerobe only (Prevotella bivi) was isolated from another dog. Cytology was inflammatory in these 7 cases with a predominance of neutrophils (n = 4), lymphocytes (n = 2), and mixed neutrophilic, lymphocytic (n = 1) inflammation.
Three of 105 dogs had cytologically detectable intracellular bacteria, but no microbial growth on aerobic, anaerobic, or Mycoplasma culture. One of these dogs had received a single dose of ampicillin with enrofloxacin the morning of the procedure. There was no recent history of antimicrobial drug administration in the remaining 2 dogs.
Risk Factors for Infection with Different Bacterial Species
Dogs from which Mycoplasma was isolated were younger than dogs from which enteric species were isolated (P < .0001, Table 1). Dogs from which B. bronchiseptica was isolated were younger than dogs from which Enterobacteriaceae, Pseudomonas, anaerobes, or Gram positive aerobic bacteria were isolated (P < .001, Table 1). Dogs from which only Mycoplasma was isolated were no more likely to be ≤1 year of age than dogs from which Mycoplasma was isolated in combination with other bacteria (1/8 versus 10/21, P = .11). Similarly, dogs from which only B. bronchiseptica was isolated were no more likely to be ≤1 year of age than dogs from which B. bronchiseptica was isolated in combination with other bacteria (6/10 versus 11/13, P = .34). Dogs from which Mycoplasma or B. bronchiseptica were isolated had a lower body weight than dogs from which other bacteria were isolated, with the exception of Pseudomonas (P < .001) (Table 1). Cough was chronic in most dogs (median 120 days, range 1–1800 days), and only 19/105 dogs (18%) had a duration of cough <1 week. At admission, mean rectal temperature was 101.6 ± 1.3°F, with temperature >102.5°F in 16/101, and mean respiratory rate was 41 ± 19 breaths per minute. There was no difference in these variables among groups of dogs from which different bacterial species were isolated.
Antimicrobial Drug Treatment
In 30 dogs with positive culture results, antimicrobial drug therapy was being administered at the time of bronchoscopy. Fourteen dogs were being treated with beta-lactams, 8 with fluoroquinolones, 5 with a combination of a beta-lactam and a fluoroquinolone, and 1 each with metronidazole, chloramphenicol, or trimethoprim-sulfamethoxazole. Eight of the 27 dogs that lacked bacteria on cytology were on antibiotics at the time of sampling (ß-lactams in 7 and a fluoroquinolone in 1).
Twelve of the 30 dogs from which Mycoplasma was isolated were being treated with antimicrobial drugs at the time of specimen collection. Five of the 9 dogs from which Mycoplasma was the only bacterial isolate were receiving treatment with a beta-lactam antibiotic. In the 21 dogs from which Mycoplasma was isolated together with other bacteria (B. bronchiseptica, Pasteurella, or enterics), 4 were being treated with beta-lactams and 3 with enrofloxacin.
Administration of beta-lactams (n = 4), fluoroquinolones (n = 3, one in combination with a beta lactam), and metronidazole (n = 1) was confirmed in 7/23 dogs from which B. bronchiseptica was isolated. Three dogs that were treated with amoxicillin-clavulanate had B. bronchiseptica isolates that were susceptible to this drug.
Antimicrobial Drug Susceptibility
Pasteurella isolates were susceptible to most antimicrobial drugs tested and Pseudomonas/Stenotrophomonas isolates were multidrug-resistant, with the exception that all were susceptible to amikacin (Table 3). More than 75% of enteric isolates were susceptible to a number of antimicrobial drugs (Table 4), while B. bronchiseptica isolates were resistant primarily to cephalosporins.
|Bacterium (Total number of isolates tested)||Bordetella bronchiseptica (25)||E. coli (18)||Klebsiella spp. (6)||Pasteurella spp. (27)||Streptococcus spp. (3)||Staphylococcus spp. (5)||Pseudomonas spp. (7)||Stenotrophomonas maltophilia (2)|
|Organism Isolated||% of Isolates Susceptible|
|Enteric organisms|| |
Lower respiratory tract infection was documented by evaluation of cytologic or microbiologic findings of BAL fluid from 105 dogs in this study, or approximately 20% of dogs that underwent bronchoscopy during the time period identified. Aerobes were isolated most often, followed by Mycoplasma and anaerobic bacteria, with isolation of multiple bacterial species in the majority (55%) of cases. Intracellular organisms were identified cytologically in most (74%), but not in all specimens.
The most common isolate from dogs with lower respiratory tract infection in this study was Mycoplasma. A high prevalence of Mycoplasma isolates in the absence of other bacterial species could be related to the common use of beta-lactam antibiotics in this group of dogs, which might select for mycoplasmal growth over that of other bacteria. Also, it is very likely associated with use of a Mycoplasma-specific culture methodology at our institution. In another study, use of Mycoplasma-specific culture media was also associated with a high prevalence of Mycoplasma isolation (65/93; 70%) in dogs with pneumonia.
The pathogenicity of Mycoplasma species in the lower respiratory tract has not been clearly established. Similar isolation rates for Mycoplasma have been reported in tracheal wash samples from healthy dogs compared with dogs with respiratory tract disease when animals >1 year of age are considered. In a previous study, dogs from which Mycoplasma was isolated responded equally well to nonspecific antibiotic therapy (trimethoprim-sulfonamides, cephalexin, penicillin) as did those that were given a drug with activity against Mycoplasma (clindamycin, tetracycline, chloramphenicol, enrofloxacin) for management of pneumonia. These studies might suggest that Mycoplasma is not a respiratory pathogen. However, in a shelter study from the UK, M. cynos was isolated from bronchial lavage samples of 10% of healthy dogs compared to 22% of shelter dogs with respiratory disease, suggesting a pathogenic effect. The role of M. cynos in the pet population has not been fully evaluated, although in the UK, serologic evidence of exposure to M. cynos was found on entry to a shelter and seroconversion increased significantly during housing, particularly in animals with respiratory disease. In a breeding facility, M. cynos was associated with lethal pneumonia in a litter of puppies. Detection of Mycoplasma as the sole isolate in both young and adult dogs here, as in other studies,[12, 17, 18] could support its role as a primary respiratory pathogen, but further investigation is required, and specific identification of mycoplasmal isolates might be needed to assess pathogenicity.
Bacteria were not visualized in samples from which only Mycoplasma was isolated, which is not surprising given the lack of staining characteristics of this bacterium. Inflammation was identified in airway fluid of all affected dogs and was primarily neutrophilic but sometimes lymphocytic. It cannot be determined whether the bacteria initiated inflammation or if bacteria collected in an inflamed environment, making it difficult to interpret the role of Mycoplasma in lower respiratory tract disease. Also, coinfection with viral organisms could enhance pathogenicity or organism accumulation. Interestingly, noninflammatory airway cytology was reported in 8/58 (14%) dogs with pneumonia that had Mycoplasma isolated, which further complicates our understanding of the role of Mycoplasma in lower respiratory tract disease.
Dogs with lower respiratory tract infection that had Mycoplasma or B. bronchiseptica isolated were lower in body weight than dogs with enteric organisms isolated and were younger than dogs with most other organisms isolated. This could be related to technique as smaller dogs were not intubated for bronchoscopy, and this could increase the risk of upper airway contamination of the sample. However, dogs with oropharyngeal contamination were excluded from the analysis and approximately 1/3 of dogs with Mycoplasma or Bordetella isolated were intubated for the procedure. Mycoplasma and B. bronchiseptica are typically associated with contagious respiratory disease, which is more common in younger dogs.[3, 14] However, almost 40% (7/18) of dogs from which these organisms were isolated in pure culture were older than 1 year of age. A previous study also detected Mycoplasma in airway specimens from older dogs with pneumonia, which indicates that Mycoplasma should be considered as a possible cause of or contributor to pneumonia in older dogs.
In this study, lower respiratory tract infection was defined by the presence of BAL fluid sepsis or substantial growth in culture, whereas a previous study used growth of >1.7 × 103 CFU/mL from a quantitative culture to diagnose infection. Quantitative cultures are not performed at our hospital, and we chose to use bacterial culture or cytology to define lower respiratory tract infection in dogs with relevant clinical signs to improve the ability to identify infection. Excluding cases with only Mycoplasma isolated, cytologic evidence of intracellular bacteria in BAL fluid was lacking in 21% of cases examined here, similar to 29% lacking bacteria in an earlier study of bronchial wash samples. This could be related to many factors, including previous or concurrent antimicrobial drug therapy that reduced bacterial numbers. Bronchial mucus, cellular disruption, excessively thick clumps of cells, or poor quality of staining might also have limited the ability to detect intracellular microbes on cytospin preparations. Additionally, failure to observe organisms could be related to specific properties of organisms such as B. bronchiseptica including sequestration of bacteria within the cilia of the respiratory tract. Given our inclusion criteria for the study along with these results, it is possible that dogs with lower respiratory tract infection were not included in this study because of both negative bacterial cultures and lack of intracellular bacteria. This study highlights the difficulty in confirming a diagnosis of lower respiratory tract infection in dogs, even when advanced testing modalities are used.
B. bronchiseptica was isolated commonly in this study (23/105 cases), as a sole isolate or in combination with other bacteria. In 3 of 4 dogs from which B. bronchiseptica was isolated, bacteria were susceptible to an antibiotic currently being administered (amoxicillin clavulanate), yet dogs had not improved clinically with treatment. In a previous study of B. bronchiseptica isolates, wide susceptibility to most antimicrobial drugs was demonstrated, including 100% susceptibility to enrofloxacin. In this study, many B. bronchiseptica species were resistant to some antimicrobial drugs tested (Tables 3 and 4). Therefore, it is difficult to make specific recommendations for treatment when Bordetella is suspected or isolated. Experimentally, nebulization with aminoglycosides will reduce bacterial numbers in the airway within 3 days, although bacterial numbers then return to pretreatment levels. Further investigation of appropriate treatment in naturally occurring disease is needed.
Anaerobes can be difficult to isolate because of the need for specialized media, loss of viability after collection when exposed to oxygen, fragile growth requirements, and a tendency to be overgrown by aerobic bacteria. Nevertheless, similar to a previous study, these microbes complicated infection in 17% of cases examined here. Anaerobes were isolated in over half of the dogs with foreign body pneumonia, but were also found in cases without foreign bodies, indicating that the lung can create an anaerobic environment in the absence of an inciting object. In a separate study, foreign bodies were also associated with anaerobic infection in approximately half of cases, Interestingly, anaerobes were the sole isolate in 4% of dogs in the current study. As in the previous study of transtracheal wash specimens, the most common anaerobe isolated in this study was Bacteroides/Prevotella, although Peptostreptococcus anaerobius was also common. Susceptibility testing is rarely performed for anaerobic bacteria because it is laborious, expensive, and has a long turnaround time; however, identification of the species present can be important for therapeutic decision making, as Bacteroides fragilis and Clostridium are less likely to be susceptible to clindamycin than to other antibiotics. Furthermore, Bacteroides fragilis often produces β-lactamase enzymes making it resistant to beta lactams such as pencillin, ampicillin, or amoxicillin.[22, 24, 25]
In this study, polymicrobial infection was common, as previously reported in adult dogs with respiratory tract infections.[2, 12] Polymicrobial infection with Streptococcus, Mycoplasma, and B. bronchiseptica was also a characteristic of lower respiratory disease in dogs <1 year of age, although a recent study of pneumonia in young dogs most commonly reported isolation of a single microbe.
A study of dogs with ventilator-associated pneumonia at our institution found a high level of antimicrobial resistance in enteric isolates. In contrast, isolates in this study of dogs with infection that developed outside of the hospital environment had similar susceptibility patterns to those determined previously from this institution, suggesting that treatment with routinely used antimicrobials drugs is likely to be effective in most dogs. Newly developed drugs such as carbapenems would appear to be rarely indicated. As might be anticipated, Pseudomonas was the most highly resistant species documented here. No cases of lower respiratory tract infection with methicillin-resistant staphylococci were identified. This is not surprising given that risk factors such as previous systemic antimicrobial use was lacking in 4/5 dogs with isolation of this organism.
This study illustrates the difficulty in confirming a definitive diagnosis of lower respiratory tract infection, even when bronchoscopy is used to obtain an airway sample. It also confirms the common occurrence of polymicrobial lower respiratory tract infections in dogs, suggesting the need to investigate multiple bacterial types when considering this diagnosis. Further research is needed to determine the pathogenicity of Mycoplasma species in dogs.
Conflict of Interest: Authors disclose no conflict of interest.
Olympus P20D, Melville NY or Pentax FG16X or FG16V, Pentax Medical Company, Montvale, NJ
Karl Storz 60003VB, Goleta, CA or Olympus BF 3C160, Melville, NY
Hardy Diagnostics, Santa Maria, CA
Anaerobe Systems, Morgan Hill, CA
UC Davis Veterinary Medicine Biological Media Services, Davis, CA
GraphPad Prism Version 5, San Diego, CA
- 5Bacterial isolates from the lower trachea of clinically healthy dogs. J Am Anim Hosp Assoc 1984;20:139–142., , .
- 7Differences in management and mortality with a bronchoalveolar lavage-based diagnostic protocol for ventilator-associated pneumonia. J Trauma Acute Care Surg 2012;72:242–246., .
- 9A Diagnostic Manual of Veterinary Clinical Bacteriology and Mycology. Davis, CA: University of California Press; 2000., , .
- 10Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Disk and Dilution Susceptibility Tests for Bacteria Isolated from Animals: Approved Standard-Third Edition. CLSI Document M31-A3. Wayne, PA: Clinical and Laboratory Standards Institute; 2008.
- 11Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing: Twenty-First Informational Supplement. CLSI Document M100-S21. Wayne, PA: Clinical and Laboratory Standards Institute; 2011.
- 18Spontaneous bronchopneumonia in laboratory dogs infected with untyped Mycoplasma. Lab Anim Sci 1990;4:625–628., , , et al.
- 22First comprehensive evaluation of the M.I.C. evaluator device compared to Etest and CSLI reference dilution methods for antimicrobial susceptibility testing of clinical strains of anaerobes and other fastidious bacterial species. J Clin Microbiol 2012;50:1153–1157., , , .
- 24ESCMID Study Group on Antimicrobial Resistance in Anaerobic Bacteria. Antimicrobial susceptibility of Bacteroides fragilis group isolates in Europe: 20 years of experience. Clin Microbiol Infect 2011;17:371–379., , .