An 80-year-old man with a remote history of a small left-hemisphere stroke lives in Florida in winter, returning home each summer. One night in Florida he awoke with flu-like symptoms, experiencing nausea and vomiting and gasping for air. His wife called 911. Emergency medical technicians noted hypoxemia and rapid respiratory rate and administered oxygen with minimal effect. Emergency department test results indicated infiltrates in the base of the right lung, elevated white blood cell count, fever, a cough with blood-tinged frothy sputum, and pulse oximetry of 85% with high-dose oxygen via a non-rebreather facemask. Aspiration pneumonia was diagnosed based upon his history of stroke, right basilar lung infiltrates, and other findings.

In the intensive care unit his condition worsened, requiring intubation and mechanical ventilation with routine sedation and chemical paralytics. He stabilized within a week; the infiltrates resolved, oxygenation improved, and he was weaned from the ventilator and extubated. The attending physician consulted with speech-language pathology, citing stroke history and the admitting diagnosis of aspiration pneumonia. A clinical examination revealed a right central facial weakness and a "gurgling vocal quality after oral trials of ice chips and puree bolus." The speech-language pathology report indicated "high aspiration risk" with oral intake described as dangerous. The physician, citing only the SLP's clinical evaluation results, consulted gastroenterology for an evaluation, and the patient underwent immediate placement of a gastrostomy because of aspiration risk. Later he was discharged to a skilled nursing facility.

Six months after this incident, the patient and his wife drove north to their hometown for the summer. He required nursing home care because his wife felt unable to manage his feeding tube regimen. He had lost 25 pounds, but otherwise was independent and asked to go home. The patient asked his family physician if he would ever be able to eat and drink again, and the physician referred him to a speech-language pathologist specializing in swallowing disorders. A clinical exam revealed a right central facial weakness, with otherwise symmetrical sensorimotor function of all structures innervated by cranial nerves V, VII, and IX-XII; functional communication for day-to-day needs with mild memory impairments; and the same post-swallow vocal quality observations reported in the inpatient evaluation.

A videofluoroscopic swallowing study revealed the following: large cervical osteophytes obstructing complete epiglottic inversion despite bolus volume or texture, causing vallecular residue of less than 25% of vallecular height, occasional shallow laryngeal penetration on four of 20 trials (two with and two without post-swallow laryngeal residue), no aspiration, otherwise intact and functional oral preparatory and transit activity, velopharyngeal closure, pharyngeal transit and clearance, upper esophageal sphincter opening and clearance, airway protection, and hyolaryngeal excursion. The patient was considered to be "back to baseline."

Treatment ensued with a goal of independence from the feeding tube. He was terrified to eat or drink. With encouragement, he gained weight and within three weeks had not used the feeding tube for 10 days. It was removed in time for Thanksgiving, which he enjoyed with his family. He was discharged from the skilled nursing facility, and the couple wife traveled back to Florida without further swallowing-related problems.

The case study above highlights the complexity of clinical decision-making and the broad range of evidence that must be reviewed to prevent adverse outcomes. One problem in this case was the early use of the diagnostic term "aspiration pneumonia," which reflects the growing focus on oropharyngeal dysphagia as a cause of adverse health consequences in patients with stroke and other neurological diseases, head and neck cancer, and other conditions. In the past 30 years an entire subspecialty has emerged. Aspiration has become the target of intervention, with methods developed to alleviate it. Now oropharyngeal dysphagia and related aspiration are recognized as a significant risk factor for morbidity and mortality after these illnesses.

In the 1990s, the number of elderly patients hospitalized with aspiration pneumonia in the United States increased four to five times more rapidly than other types of pneumonias without a logical explanation for its disproportionate growth (Baine, Yu, & Summe, 2001). Were more people aspirating or was there a significant increase of idiopathic aspiration? There was no corresponding rise in the incidence of stroke, neurological disease, or illnesses that predispose a person toward aspiration.

Aspiration contributes to the pathogenesis of dysphagia-related pneumonia (Marik, 2001), but only as one of many pneumonia risk factors, and sometimes not even the greatest. There are two sources of aspiration—the mouth and the stomach. Interventions are often directed at the complete eradication of aspiration and laryngeal penetration, although both are sometimes observed in the swallow of healthy individuals (Robbins, Coyle, Rosenbek, Roecker, & Wood, 1999). Aspiration also occurs during sleep in many healthy individuals. We prescribe postural maneuvers, modify diets, and thicken liquids to eliminate aspiration, and patients sometimes receive feeding tubes because of aspiration risk. Often clinicians forget that other pneumonia risk factors outweigh the potential harm of aspiration, and may use interventions that could pose greater risks to the patient than the swallowing disorder itself.

In the 1990s, most SLPs viewed aspiration only as related to oropharyngeal dysphagia; even now, when aspiration is suspected, our minds jump directly to dysphagia. In stroke, mortality is six times higher if the patient develops pneumonia after stroke onset (Katzan et al., 2003). Formal screening of swallow function has been shown to reduce inpatient pneumonia risk significantly after stroke (Hinchey et al., 2005). The recent report by the Agency for Healthcare Research and Quality (AHRQ) drew greater attention to this issue, spawning stroke dysphagia screening protocols; now nearly all patients with the preliminary diagnosis of "aspiration pneumonia" (with or without substantiating case evidence) are referred to hospital speech-language pathology departments (Doggett et al., 2001).

To manage a case accurately, the SLP must differentiate among the signs and markers that typically accompany the adverse medical outcomes of dysphagia and determine which are more likely to have been caused by dysphagia and which are more likely due to other conditions. SLPs typically are charged with determining whether their patients have dysphagia-related aspiration pneumonia (DAP), non-dysphagia-related aspiration pneumonia (NDAP), community-acquired pneumonia, health care-associated (nosocomial) pneumonia, or aspiration pneumonitis (the correct diagnosis in our sample case). Physicians often take SLPs' written reports and recommendations as direct evidence supporting medical and surgical interventions, whether or not that evaluation was sufficiently thorough in the analysis of case information outside of the SLP's traditional comfort zone. This knowledge of pulmonary physiology and diseases is not covered in SLPs' training, but clinicians must embrace this knowledge to assess accurately the risk that dysphagia (or something else) poses to their patients.

The Chicken or the Egg?  

When an SLP examines a patient diagnosed with aspiration pneumonia and identifies dangerous swallowing function, the clinician's observations often "prove" the hypothesis that dysphagia caused the disease. But it may be unclear whether dysphagia caused the disease, or if the disease caused the dysphagia. A hospitalized patient with pneumonia looks very ill—clinicians may see lethargy, acute encephalopathy associated with hypoxemia and medication side effects, and malnutrition due to reduced intake. These observations may cause them to suspect aspiration risk.

When the effects of medication (including potent sedatives and paralytics used during mechanical ventilation) subside, the patient with acute pneumonia—"obviously dysphagic," hypoxemic, febrile, dyspneic, and possibly malnourished with a nasogastric tube in place—is rolled into radiology and undergoes endofluoroscopy. In that situation, the SLP likely will see very poor swallowing function, especially if it already was predicted.

What the clinician sees, however, often is not the patient's condition when the pneumonia was contracted. Certainly obvious risks exist because of acute decompensation, but the clinician cannot assess with certainty the causal condition. Is the clinician seeing the chicken (dysphagia caused the aspiration pneumonia) or the egg (pneumonia has caused acute reversible dysphagia)? It is possible that the truth is more complex—that dysphagia has caused pneumonia that has acutely worsened pre-existing dysphagia.

When an SLP reports "aspiration"—or worse, "aspiration risk"—in the chart, the physician often responds by recommending a surgically placed feeding tube. This placement often occurs whether or not the acute and potentially rapidly reversible nature of the patient's current condition has been considered, or whether the increased risk of a feeding tube favoring aspiration of stomach contents was assessed. As in our sample case, within days there was no dysphagia because its underlying cause had subsided. But now our patient has a feeding tube. The SLP's report must reflect the natural history of the cause of the dysphagia symptom and its prognosis—if due to acute illness that causes global decompensation with a short recovery period, the prognosis for short-term recovery is likely very good. These realities need to be addressed in the recommendations.

Feeding Tubes and Aspiration Sources  

Reducing the risk of aspiration associated with dysphagia is commendable—unless the risk is shifted to another and possibly worse aspiration source. There is substantial evidence that the condition of some individuals who aspirate when swallowing worsens when a feeding tube is placed (Finucane, Christmas, & Travis, 1999; Gomes et al., 2003; Goodwin, 1996; Langmore et al., 1998). Placement of feeding tubes in patients with documented gastroesophageal reflux or other upper digestive conditions—even a patient with dysphasia—may  dramatically increase the risk of aspiration from the stomach far beyond the patient's aspiration risk related to oropharyngeal dysphagia (Centers for Disease Control and Prevention, 2003; Ciocon et al., 1988; Dziewas et al., 2004).

Similarly, as in our sample case, the aspiration of gastric contents during emesis can be mistakenly overlooked, though aspiration pneumonitis (the acute lung injury caused by aspiration of caustic gastric contents) has been reported in the literature for decades (Marik, 2001; Mendelson, 1946). Understanding the benefits and risks of feeding tubes is essential in identifying which intervention may be the safest for the patient.

The Role of Pathogens  

The pneumonia syndrome is an inflammation of alveoli and associated airways caused by pathogenic (disease-causing) microorganisms. The healthy respiratory system tolerates many microorganisms; the damaged system does not. Underlying pulmonary diseases, an immunocompromised condition, and active smoking increase the risk that aspiration will produce significant pneumonia (Langmore et al., 1998; Mokhlesi et al., 2002; Stein et al., 1990). The pathogen invades and draws its nutrition from the area of lung it has invaded, causing inflammation, and the infection spreads through bacterial reproduction and by coughing (which projects the pathogens upstream in the airways to be inhaled into other pulmonary segments). This pathogen-caused inflammation (an infection) causes thickening, toughening, and reduced compliance of alveolar membranes, leading to respiratory distress. The healthy immune system increases production of white blood cells to combat infection (leukocytosis); the immunocompromised patient's response, however, is poor.

Fever, another sign of inflammation, persists while the infection progresses, though in cases of acute inflammation (as in the aspiration pneumonitis in our sample case) it may subside quickly. The alveolar membrane and associated pulmonary capillaries become inflamed and excessively permeable, allowing fluid to leak into the alveoli from the blood. This leakage, combined with the pathogens, white blood cells, and other debris and bacterial waste, combine to fill the alveolar airspaces (pulmonary infiltrates) progressively, reducing surface area available for respiratory gas exchange. Shortness of breath ensues as the patient's respiratory rate increases to maintain oxygen and carbon dioxide exchange, increasing the potential loss of respiratory-swallow coordination. An infection not treated soon enough will cross from the alveoli through the leaky capillary membranes into the blood to travel to other organs, producing septicemia (or sepsis) with multi-organ infection and failure.

Oropharyngeal secretions contain many microorganisms, ordinarily not pathogenic. Dysphagia can enable aspiration of oropharyngeal secretions containing these organisms. The maintenance of oral hygiene is supported by scores of studies demonstrating that a) patients with teeth have a higher oral pathogen load, and b) high pathogen loads in saliva of patients with poor oral hygiene raise the risk of pneumonia (Adachi et al., 2007; Azarpazhooh & Leake, 2006; Langmore et al., 2002). Does evidence of oral pathogens in the sputum of a patient with pneumonia provide conclusive evidence that dysphagia caused the pneumonia? Maybe not. Oral pathogens in the sputum may come from the stomach, the mouth, or neither.

Numerous studies have demonstrated aspiration of gastric contents in healthy and diseased patients (Metheny et al., 2006; Ravelli et al., 2006; Tobin et al., 1998). Deacidification of gastric contents with acid-suppressing drugs used to treat gastroesophageal reflux (e.g., anti-reflux medications) creates an environment in the stomach that allows some swallowed microorganisms to thrive, which has been associated with an increased pneumonia risk (Eurich et al., 2010; Hauben et al., 2007; Laheij et al., 2004) and mortality (Herzig et al., 2009; Rantanen & Salo, 1999). Therefore, a patient receiving acid-suppression therapy who has normal swallowing function and gastroesophageal reflux and who swallows oral pathogens into the stomach may aspirate oral pathogens from the stomach. This patient can develop aspiration pneumonia without dysphagia and get very sick and temporarily dysphagic. An SLP may then attempt to isolate, and rule in or out, dysphagia-related aspiration as a cause of the pneumonia.

The bacteriology of a pulmonary infection taken from a sputum sample can be useful in isolating a probable cause and guiding antibiotic selection, and in attributing or refuting dysphagia as a cause (Gotfried, 2001; Marik, 2001). SLPs must be careful, however, when interpreting sputum cultures.

A sputum sample obtained at bedside basically involves a "spit-into-the-cup" method that contaminates pulmonary specimens with oral bacteria. "Normal oral flora" growth found in a sputum specimen obtained at the bedside is equivocal as evidence of dysphagia-related aspiration (but not to other sources of pneumonia). Sputum samples obtained through invasive bronchoscopic methods that prevent oral contamination are extremely sensitive and specific to the exact pathogen found in the lung, but these methods are risky, invasive, and reserved for patients who are critically ill, intubated, or tracheostomized. "Normal oral flora" found in protected specimens obtained from the lung may produce reliable information (unless acid suppression is in place).

Pneumonias Are Not All the Same  

Aren't all pneumonias "aspiration pneumonias" because a pathogen enters the airway? No. Aspiration of liquid or solid matter does not occur during normal breathing. Aspirated material is drawn to gravity-dependent portions of the respiratory system (i.e., the bottom of the lungs). Yes, the right main stem bronchus is more vertically positioned in most adults than the left; hence the attribution of "right lower lobe pneumonia" to aspiration and dysphagia.

Airborne pathogens are dependent on airway diameter and airflow direction and velocity (the Bernoulli effect). Annual pneumonia vaccinations are given to health care workers and at-risk individuals to protect them from pathogens that cause community-acquired pneumonia. These pathogens (streptococcus pneumoniae, pneumococci, hemophilus influenzae, and others) typically are airborne although they also can colonize the oropharyngeal mucosa. Likewise, nosocomial pathogens such as staphylococcus aureus, pseudomonas and proteus species, and others that live in the nooks of hospitals and skilled nursing facilities waiting to colonize a host, typically infect wounds; they also can colonize aerodigestive mucosa and be aspirated.

Reaching the True Diagnosis 

Determining whether aspiration pneumonia is or is not a true diagnosis sometimes leads to unexpected outcomes. In our sample case—a true story—the patient was misdiagnosed with aspiration pneumonia. He developed acute chemical pneumonitis due to aspiration of caustic gastric contents during emesis. His swallow function was not normal, but it was not pathological. The clinical sign—"gurgle after the swallow"—was caused by pharyngeal residue due to cervical osteophytes, a baseline condition. But it led to a feeding tube.

The mismanagement of this patient was caused by an erroneous diagnosis, which led to an unnecessary invasive, restrictive, unpleasant, and painful procedure. To avoid this outcome, medical SLPs are obligated to know the natural history of aspiration pneumonitis and aspiration pneumonia, the numerous risk factors for each and how each affects each patient, and the ways in which aspiration pneumonia and mimicking conditions can be caused.

The skill with which SLPs can gain and use this knowledge predicts the precision of our clinical
decision-making and in turn, the advancement of medical speech-language pathology. In most cases, risk cannot be eliminated—only lowered—by one intervention, and perhaps reduced further by another. Although none of the interventions available eliminates risk to our patients, our aim should be to identify which intervention provides the lowest risk of an adverse outcome, balanced with expectation of a reasonable benefit.