Aspiration Pneumonia: Current Concepts and Approach to Management

Aspiration Pneumonia: Current Concepts and Approach to Management Aspiration pneumonia usually presents in patients with underlying disease that predisposes to host defense impairment. Identification of causative pathogens and targeting high-risk patients are critical to effective management. journalArticle Hugh A. Cassiere, MD, Winthrop University Hospital Dr. Cassiere is Medical Director of the Cardiac Surgical Intensive Care Unit, and Attending Pulmonary and Critical Care Medicine Physician, Department of Thoracic and Cardiovascular Surgery, Winthrop University Hospital, Mineola, NY. Cassiere HA. Aspiration Pneumonia: Current Concepts and Approach to Management. MedGenMed 1(3), 1999 [formerly published in Medscape Pulmonary Medicine eJournal 2(1), 1998]. Available at: http://www.medscape.com/viewarticle/408725 Medscape Medscape General Medicine [TM] 01/20/1999 1 3 <H3>Abstract and Introduction</H3><H4>Abstract</H4> Aspiration of pathogens from a previously colonized oropharynx is the primary route by which organisms gain entrance to the lungs. Like other respiratory tract infections, aspiration pneumonia most commonly manifests in patients with underlying disease that predisposes to host defense impairment. Conditions which compromise host immunity to aspirates include diabetes mellitus, congestive heart failure, COPD, malnutrition, renal failure, and malignancy. The clinical response to aspirated material is dependent on the interplay between the characteristics of the aspirate and those of the host. The majority of large volume and particulate aspirations are comprised of vegetable matter which can mechanically obstruct the lower airways and cause atelectasis, stagnation of secretions, and thus an increased risk of infection. In addition, particulate-matter aspiration can be contaminated by bacteria because oral secretions are often colonized by potentially pathogenic organisms. The microbiology of aspiration pneumonia has not changed considerably over the last few decades, and is intimately tied to the predominantly anaerobic flora of the oropharyngeal cavity. Aerobic organisms are found as either primary pathogens or as co-infectors. The recent report of a more virulent aerobic pathogen profile in ICU patients with early aspiration pneumonia, however, suggests that such severely ill individuals may have a different bacteriology. The two primary approaches to the management of aspiration pneumonia are antibiotics and supportive care. Antibiotic therapy should be based on an assessment of the severity of illness, where the infection was acquired (community versus hospital), and the presence or absence of risk factors for gram-negative rod colonization.<H4>Case Presentation</H4> MS is a 58-year-old male with a history of noninsulin-dependent diabetes mellitus who is two days post uncomplicated coronary artery bypass graft (CABG). He remains in the cardiac surgical step-down unit because of persistent bradycardia requiring external pacing via epicardial wires. After turning off the external pacemaker accidentally, the patient switches the device back on. A rhythm strip taken after the pacemaker was turned back on is shown in Figure 1. During DC cardioversion from the resulting arrhymthia (torsades de pointes), MS has a witnessed episode of aspiration.<center><img src="art-mrc3075.fig1.jpg" width="400" height="331" BORDER="1"></center><blockquote> Rhythm strip showing torsades de pointes, which occurred after the patient accidentally switched off his external pacing device and then reactivated it. DC cardioversion was required.</blockquote> The patient's respiratory status worsened over the next few hours requiring endotracheal intubation and mechanical ventilation. His pulse rate was 110 and his respiratory rate was 14. His blood pressure was 100/50 mm Hg and his temperature was 38.1deg.C (100.6° F). Significant findings on physical included poor dentition, clean and dry surgical incisions, and scattered rhonchi in the anterior lung fields. Laboratory data: hemoglobin, 10.5 g/dL; WBC count, 21,000/mm3; potassium, 3.2 meq/mL and creatinine, 2.1 mg/dL. For analysis of arterial blood gases, findings were: pH 7.37, pO2 63 mmHg, pCO2 41 mmHg, and HCO3 23 meq/mL on FiO2 of 60%. Tracheal aspirate samples obtained after intubation grew K pneumoniae (Fig. 2)and S intermedius. Chest radiograph taken several hours after the cardioversion and aspiration episode showed fluffy bilateral infiltrates, which resolved after several days of antibiotic therapy (Figs. 3,4). The patient was continued on antibiotics and was extubated several days later with full recovery.<center><img src="art-mrc3075.fig2.jpg" width="400" height="260" BORDER="1"></center><blockquote> Pathology slide of tracheal aspirate performed after intubation grew K pneumoniae</blockquote> <center><img src="art-mrc3075.fig3.jpg" width="458" height="331" BORDER="1"></center><blockquote> Chest radiograph taken several hours after aspiration episode showed fluffy bilateral infiltrates.</blockquote> <center><img src="art-mrc3075.fig4.jpg" width="444" height="322" BORDER="1"></center><blockquote> Chest radiograph taken after several days of antibiotic therapy shows resolution of the pulmonary infiltrates.</blockquote> <H4>Introduction</H4> Pathogenic organisms can invade and infect the lung by four routes : aspiration, inhalation, hematogenous seeding, and contiguous spread. Aspiration of pathogens from a previously colonized oropharynx is the primary pathway by which organisms gain entrance to the lungs, and therefore, in a broad sense, most pneumonias, are aspiration-related. However, when the term "aspiration pneumonia" is used, it is meant to refer to the development of a radiographic infiltrate in a patient with either a witnessed episode of gross aspiration or risk factors for aspiration.Everyone aspirates...Aspiration pneumonia, like other respiratory tract infections, usually occurs in patients with underlying disease that predisposes to host defense impairment. Aspiration pneumonia most commonly occurs in post- stroke or postgastrectomy patients, and in those with dysphagia, gastroesophageal reflux, xerostomia, periodontal disease, altered sensorium, or underlying serious illness. Normal adults, however, commonly aspirate while sleeping without any obvious health effects. In addition to host characteristics, more frequent aspiration, larger aspirate volume, and acidic aspirate pH and hypertonicity increase the risk of developing pneumonia. The finding of infiltrate on plain radiograph following aspiration may represent an infectious process or a noninfectious inflammatory response, resulting in pulmonary leukocyte sequestration. Many noninfectious infiltrates resolve, but in 25% to 50% of cases, even noninfectious infiltrates seen on plain radiograph progress to infectious pneumonia or fulminant acute lung injury.[1]...But pneumonia is more likely in "aspirators." Croghan and colleagues[2] examined nursing home patients evaluated for aspiration with videofluoroscopy and found 50% of known aspirators developed pneumonia over a 12-month period, compared with 12.5% of patients not known to aspirate. In this study, the volume of the aspirate was directly related to pneumonia risk: 43% of patients with mild aspiration (thin liquid aspirations) developed pneumonia, compared with 67% of patients with severe aspiration (all consistencies aspirated with almost every swallow). Bynum and coworkers,[3] evaluated 50 patients with significant aspiration. Sixty-two percent showed rapid clearing of infiltrates, while 12% developed ARDS (adult respiratory distress syndrome).[3] The remaining 26% of patients developed infectious pneumonia, often after initial improvement.<H3>Increased Risk-Host Characteristics</H3> Risk factors for acquiring aspiration pneumonia are numerous and complex. One conceptual approach is to divide risk factors into host factors and aspirate factors (Table I). Host factors may be further divided into host defenses and increased risk for episodes of aspiration. In general, it is the combination of disease-related immune impairments and exposure to large numbers of aspirated bacteria that predisposes to pneumonia.Underlying disease: Mechanisms of decreased host defense. Underlying disease is one of many host factors that predispose to aspiration pneumonia. Conditions that weaken the host response to aspirated material include diabetes mellitus, congestive heart failure, malnutrition, chronic obstructive pulmonary disease (COPD), renal failure, and malignancy. Diabetes mellitus, as was found in our patient, is associated with neutrophil dysfunction, including diminished chemotaxis and impairment of phagocytosis. Patients with congestive heart failure can have impaired clearance of pneumococci and staphylococci from the respiratory tract as a result of pulmonary edema. If liver disease is complicated by cirrhosis, atelectasis from pleural effusions and/or ascites can result in stagnation of secretions and increased risk of infection. Cirrhosis is also associated with diminished leukocyte chemotaxis in response to inflammation, depressed complement levels, and defects in cellular immunity. Renal failure can increase the rate of gram-negative and Staphylococcus aureus colonization of the oropharyx, and can lead to complement deficiency, impaired cellular immunity, and, in animal models, decreased clearance of staphylococci and Pseudomonas aeruginosa along with an increase in buccal cell binding capacity for gram-negative bacteria. Neoplasms can affect host defenses either directly or as a consequence of therapy. For example, neutropenia, the presence of endobronchial obstruction, and impaired cough reflex, are common in cancer patients and can all predispose to pneumonia.[4] In addition to the specific mechanisms previously indicated, cell surface fibronectin has been shown to prevent the adherence of gram-negative rods to receptors on oropharyngeal cells in normal hosts. However, in patients with underlying illness, cell surface fibronectin is cleaved off, and receptors to gram-negative rods are exposed. Xerostomia has recently been shown to predispose to aspiration pneumonia, most likely because of increased bacterial concentration in the saliva. Normally, saliva flushes the oral cavity and maintains a bacterial level of less than 7 x 108 bacteria/mL . In addition, patients are at risk for gingivitis. These two factors can result in bacterial counts that are higher than normal. When a patient with xerostomia aspirates, the lower respiratory tract is exposed to larger numbers of bacteria than normal, and aspiration pneumonia can result.[5]Increased frequency of aspiration. Host factors that predispose to aspiration are the result of neurologic, mechanical, and/or contractile dysfunction of the esophagus and upper airway (Table II). These factors include stroke, dysphagia, gastroesophageal reflux, altered sensorium, feeding tube placement, and the postgastrectomy state.[5-10] Specific issues to consider when evaluating ICU patients for aspiration risk, include patient position, site of enteral feeding, volume of gastric contents (higher volume has greater risk), and size of any feeding tube. Studies have suggested a reduced risk of aspirating gastric contents in semi-erect patients, in those whose feeding tubes are in the small bowel, and in those with small-bore feeding tubes.<H3>Increased Risk -- Aspirate Characteristics</H3> Only 25% to 50% of all aspirations progress to pneumonia. Once material has gained access to the airways, the clinical response depends upon the interplay between characteristics of the aspirate and those of the host. If the aspirate is small in volume, but highly contaminated with bacteria, then even relatively strong host defenses may be overwhelmed and pneumonia can result. If the aspirate is large in volume, but small in contagion, then pneumonia will result only if the aspirated organisms are highly virulent or the host defenses severely compromised . Even patients who aspirate noninfectious material may progress to pneumonia as a result of the lung injury caused by noxious aspirate material.[11-15]Bacterial titers high in periodontal disease. Most episodes of aspiration involve oropharyngeal flora, which can achieve extremely high concentrations, especially in the presence of periodontal disease. While normal saliva has 108 organisms/mL, saliva from a patient with gingivitis may contain 1011 organisms/mL. Anaerobes predominate in the oral flora of patients with periodontal disease. The finding that the patient presented had poor dentition on physical examination was indicative of periodontal disease.Acidic aspirates. Aspiration of material with high hydrogen ion concentrations may result in chemical pneumonitis. The initial pulmonary infiltrate following such an insult is typically noninfectious, and characterized by a predominance of neutrophils. The magnitude of lung injury is directly related to the volume and the hydrogen ion concentration of the aspirated material. In animal models of aspiration, acid pneumonitis does not occur unless pH is less than 2.5. The resulting damage renders the mucosal barrier of the lower respiratory tract incompetent -- new sites for bacterial binding are created.Vegetable and particulate aspirates. A number of studies[1,3,11] have shown that the majority of large volume and particle aspirations are comprised of vegetable matter which can mechanically obstruct the lower airways and cause atelectasis, stagnation of secretions and thus an increased risk of infection. In addition, particulate-matter aspiration can be contaminated by bacteria because oral secretions are often heavily colonized by potentially pathogenic organisms.<H3>Aspiration Syndromes</H3> Aspiration of material into the tracheobronchial tree can have clinical consequences, ranging from no evident reaction to a reaction involving ARDS, respiratory failure, and death. Three mutually unexclusive aspiration syndromes -- irritant-toxic type, the inert-nontoxic type, and the infectious type -- may result from any aspiration event, depending on the nature of material entering the tracheobronchial tree. The irritant-toxic syndrome is caused by aspirating acidic liquids (pH < 2.5) and/or fine particulate material, resulting in acute pneumonitis or ARDS. The inert-nontoxic syndrome results from the aspiration of large particulate matter and/or large volumes of fluid. The clinical response to this form of aspiration ranges from chronic respiratory complaints, such as cough and wheezing, to atelectasis, or if the aspiration is massive, sudden death. Infectious aspiration syndromes result from the entry of potentially pathogenic organisms into the lower airways. While pneumonia is part of the infectious aspiration syndrome, other infectious syndromes can result, including lung abscess, necrotizing pneumonia, and empyema. Because community-acquired aspiration pneumonia usually involves anaerobic bacteria, aspiration pneumonia should be viewed as part of a continuum that can progress to cavitation (lung abscess), or even empyema formation. If aspiration pneumonia is not treated, necrotizing abscess formation may follow in 8 to 14 days, usually in a peripheral location, and is characterized by the expectoration of putrid sputum. When anaerobic lung infection occurs, the pleura is involved in as many as half of all patients. Mendelson's syndrome (acid pneumonitis), is an example of irritant-toxic lung injury, and usually follows a witnessed episode of aspiration of acidic gastric fluid. Initially there is an asthmatic-like reaction, but within 2 hours the patient may progress to dyspnea, cough with nonpurulent sputum, bronchospasm, bilateral lower lobe infiltrates, hypoxemia, and decreased lung compliance. In some patients, ARDS and death may result. Mendelson's original description of this syndrome documented no deaths, although subsequent investigators[3,11] have reported mortality rates as high as 70%.<H3>Microbiology</H3> Most aspiration pneumonias are caused by anaerobic organisms originating from the oropharynx and are usually polymicrobial, with at least two anaerobic organisms, though sometimes there is a mixture of aerobic and anaerobic pathogens. The microbiology of aspiration pneumonia is intimately tied to the flora of the oropharyngeal cavity. Under normal circumstances the oral cavity is inhabited by 108 bacteria/mL of saliva, with a predominance of anaerobic organisms. Individuals with poor dental hygiene and gingivitis can have anaerobic bacterial levels of up to 1011 organisms/mL of saliva. Patients with prolonged hospitalization and underlying illness may become colonized by enteric gram-negative bacilli. The microbiology of aspiration pneumonia has not changed much over the last few decades, although the taxonomy of some of the involved pathogens has. For example, some organisms originally classified in the Peptostreptococcus genus have now been reclassified as Streptococcus (ie, S intermedius. ). A number of studies in the 1970-80s[1,11,15-18] reported anaerobic streptococci, Fusobacterium nucleatum, and Prevotella melaninogenicus (formerly classified in the genus Bacteroides) as the three major pathogens associated with development of aspiration pneumonia. It was once thought that Bacteroides fragilis was a significant causative pathogen in anaerobic lung infections, although recent data make this seem less likely[19] . Aerobic bacteria are found as either primary pathogens (in approximately 10% of cases ) or as co-infectors (in approximately 40% of cases ) and include Streptococcus spp, Staphylococcus aureus, Klebsiella pneumoniae, Escherichia coli, Enterobacter cloacae, and P aeruginosa. A more recent study examining early aspiration pneumonia in ICU patients has documented a more virulent aerobic pathogen profile.[19] In this series no anaerobes were isolated, but in those patients who had community-acquired early aspiration pneumonia, Streptococcus pneumoniae, S aureus, E coli, E cloacae, Haemophilus influenzae, Streptococcus viridans, and P aeruginosa were isolated alone or in combination. In those patients who aspirated in-hospital, S aureus, H influenzae, Serratia marcescens, Morganella morgannii, Candida albicans, K pneumoniae, P aeruginosa, and Proteus mirabilis were isolated alone or in combination by a protected specimen brush. These data suggest that severely ill patients with underlying medical diseases may have a different bacteriology than those with nonsevere aspiration pneumonia.<H3>Treatment</H3> The two primary approaches to the management of aspiration pneumonia are antibiotics and supportive care. In the setting of a witnessed or suspected aspiration, antibiotic therapy should be started if an infiltrate is present. If the infiltrate clears within 24 to 48 hours, it was likely due to a noninfectious pneumonitis, and therapy can be discontinued. If no infiltrate is present, therapy can be withheld, provided that serial chest radiographs are followed for 48 hours. Antibiotic therapy for patients with aspiration pneumonia should be based on an assessment of severity of illness (severe or nonsevere infection; Table III), where the infection was acquired (community versus nosocomial), and the presence or absence of risk factors for gram-negative rod colonization . If pneumonia is acquired in the community, then an anaerobic pathogen is more likely causative than if the infection was acquired in-hospital. In this circumstance, the antibiotic regimen should be directed against the oral anaerobes, including anaerobic streptococci, F nucleatum, and P melaninogenicus. Patients should be treated empirically without collecting sputum for culture by expectoration or invasive aspiration. The initial empiric antibiotic used in this case may be clindamycin, or penicillin alone, or a combination of penicillin and metronidazole (Table IV). Some data[18,19] show a lower failure rate in patients treated with clindamycin compared to penicillin for anaerobic pleuropulmonary infection, suggesting a primary role for clindamycin in this process . Recent data from patients with community-acquired lung abscess showed resistance (to anaerobes) rates to penicillin, metronidazole, and clindamycin of 21%, 12%, and 5%, respectively, again supporting a primary role for clindamycin in anaerobic lung infections.[1,15-19] This antibiotic regimen needs to be modified if the patient has severe infection, hospital-acquired aspiration, or risk factors for gram-negative colonization, which include malnutrition, severe illness, coma, intubation, diabetes, prior surgery, lung disease, renal failure, prior antibiotic use, hypotension, cigarette smoking, prolonged hospitalization. In these settings , the likelihood of infection with a virulent gram-negative bacillus or other aerobic organisms is greater and therefore additional antibiotic coverage is required. Sputum or tracheal aspirate may be helpful in identifying high-risk pathogens, such as P aeruginosa in intubated patients. After empiric therapy is started, culture results can be useful in determining whether infection is P aeruginosa-induced. In this patient population, clindamycin plus an antibiotic with activity against gram-negative bacilli is the recommended approach to management but a single antibiotic that has activity against anaerobes and gram-negative bacteria is adequate (Table IV). If risk factors for infection with P aeruginosa are present we recommend empiric antibiotic therapy with a dual anti-pseudomonal combination until culture results are known. <a name=""><h3>Table I - Risk Factors for Aspiration Pneumonia</h3></a> <blockquote><center><TABLE border=1 cellpadding=3><tr align=left><th>Host Risk Factors</th><th>Aspirate Risk Factors</th></tr><tr valign=top><td>Underlying serious illness Altered sensorium Stroke Dysphagia Gastroesophageal reflux Postgastrectomy Xerostomia Feeding tube Periodontal disease</td><td>Fluid pH < 2.5 Large particles Large volume (mL/kg) Hypertonic fluid Bacterial contamination</td></tr></table></center></blockquote> <a name=""><h3>Table II - Conditions That Increase the Risk of Aspiration</h3></a> <blockquote><center><TABLE border=1 cellpadding=3><tr align=left><th>Neurological</th><th>Mechanical</th><th>Contractile</th></tr><tr valign=top><td>Unconsciousness Laryngeal nerve damage Advanced age Acute stroke Pseudobulbar palsy Seizures Cardiac arrest General anesthesia Hypoglycemia Alcoholism</td><td>Obesity Head & neck surgery Bowel obstruction Abdominal surgery Enteral feeding Pregnancy Endotracheal intubation Tracheostomy</td><td>Gastroesophageal reflux Diabetic gastropathy Critical illness Trendelenburg position Protracted vomiting</td></tr></table></center></blockquote> <a name=""><h3>Table III - Defining Characteristics of Severe Aspiration Pneumonia</h3></a> <blockquote><center><ul><li>Respiratory rate > 30 breaths/min <li>Need for mechanical ventilation <li>Chest radiography findings: <ul><li>50% increase in the infiltrate in 48 hours <li>Bilateral multilobar involvement </ul><li>Presence of shock <li>SIRS (systemic inflammatory response syndrome) or Vasopressors to support blood pressure <li>Severe lung injury (PaO2/FiO2 ratio < 250 mmHg) <li>Urine output < 20 cc/hr <li>Acute renal failure requiring dialysis </ul></center></blockquote> <a name=""><h3>Table IV - Initial Antibiotic Regimen for Aspiration Pneumonia</h3></a> <blockquote><center><TABLE border=1 cellpadding=3><tr valign=bottom><th align=left>Community-Acquired Infection, Nonsevere</th><th align=left>Hospital or Severe Community-Acquired INFECTION</th></tr><tr><td colspan=2>Oral route</td></tr><tr valign=top><td>Penicillin Penicillin plus metronidazole Clindamycin Amoxicillin/clavulanate</td><td>NA</td></tr><tr><td colspan=2>Intravenous route</td></tr><tr valign=top><td>Imipenem* Penicillin Penicillin plus metronidazole Clindamycin Ampicillin/sulbactam Ticarcillin/clavulanate</td><td>Clindamycin plus quinolone* Clindamycin plus aminoglycoside* Ampicillin/sulbactam Cefoxitin Piperacillin*</td></tr></table><TABLE><tr valign=top><td>*</td><td>If any of the following factors are present, P aeruginosa involvement is possible: Prior antibiotic use, prolonged hospital course, and/or severe pneumonia. If P.aeruginosa is suspected, dual anti-pseudomonal therapy should be initiated.</td></tr></table></center></blockquote> <OL><li>Bartlett JG, Gorbach SL: The triple threat of aspiration pneumonia. Chest 68:560-566, 1975. <li>Croghan JE, Burke EM, Caplan S, et al: Pilot study of 12-month outcomes of nursing home patients with aspiration on videofluoroscopy. Dysphagia 9:141-146, 1994. <li>Bynum LJ, Pierce AK: Pulmonary aspiration of gastric contents. Am Rev Respir Dis 114:1129-1136, 1976. <li>Cassiere HA, Niederman MS: New etiopathogenic concepts of ventilator-associated pneumonia. Sem Respir Infect 11:13-23, 1996. <li>Terpenning M, Bretz W, Lopatin D, et al: Bacterial colonization of saliva and plaque in the elderly. Clin Inf Dis 16:S314-316, 1993. <li>Holas MA, DePippo KL, Reding MJ: Aspiration and relative risk of medical complications following stroke. Arch Neurol 51:1051-1053, 1994. <li>Holtsclaw-Berk SA, Berk SL, Thomas CT, et al: Gastric microbial flora in patients with gastrointestinal disease. South Med J 77:1231-1233, 1984. <li>Kidd D, Lawson J, Nesbitt R, et al: Aspiration in acute stroke: A clinical study with videofluoroscopy. Q J Med 86:825-829, 1993. <li>Martin BJ, Corlew MM: The association of swallowing dysfunction and aspiration pneumonia. Dysphagia 9:1-6, 1994. <li>Marumo K, Homma S, Fukuchi Y: Postgastrectomy aspiration pneumonia. Chest 107:453-456, 1995. <li>Wynne JW, Modell JH: Respiratory aspiration of stomach contents. Ann Intern Med 87:466-474, 1977. <li>Kikuchi R, Watabe N, Konno T, et al: High incidence of silent aspiration in elderly patients with community-acquired pneumonia. Am J Respir Crit Care Med 150:251-253, 1994. <li>Knight PR, Druskovich G, Tait AR, et al: The role of neutrophils, oxidants, and proteases in the pathogenesis of acid pulmonary injury. Anesthesiology 77:772-778, 1992. <li>Folkesson HG, Matthay MA, Hebert CA, et al: Acid aspiration-induced lung injury in rabbits is mediated by interleukin-8-dependent mechanisms. J Clin Invest 96:107-116, 1995. <li>Lorber B, Swenson RM: Bacteriology of aspiration pneumonia. Ann Intern Med 81:329-331, 1974. <li>Bartlett JG, Gorbach SL: Treatment of aspiration pneumonia and primary lung abscess. Penicillin G vs clindamycin. JAMA 234:935-937, 1975. <li>Bartlett JG, Gorbach SL, Finegold SM: The bacteriology of aspiration pneumonia. Am J Med 56:202-207, 1974. <li>Cesar L, Gonzalez C, Calia FM: Bacteriologic flora of aspiration-induced pulmonary infections. Arch Intern Med 135:711-714, 1975. <li>Mier L, Dreyfuss D, Darchy B, et al: Is penicillin G an adequate initial treatment for aspiration pneumonia? A prospective evaluation using a protected specimen brush and quantitative cultures. Inten Care Med 19:279-284, 1993. </ol>