Nicholas H. Fiebach
Respiratory tract infections are the most common acute illnesses in the United States and in the industrialized world. These infections are the most frequent causes of absences from school or work. Most upper respiratory infections (URIs) are self-diagnosed and self-treated and do not come to the attention of a clinician. Lower respiratory infections (LRIs) may be minor or amenable to ambulatory treatment, but they also represent one of the leading causes of hospitalization and death in this country. The cost of respiratory tract infections in lost productivity and expenditures for treatments, including over-the-counter (OTC) remedies, is estimated to be billions of dollars.
Upper Respiratory Tract Infections
Common Cold
The common cold, or coryza, is a mild, self-limited syndrome caused usually by viral infection of the upper respiratory tract mucosa and characterized by one or more of the following symptoms: nasal discharge and obstruction, sneezing, sore throat, cough, and hoarseness.
Epidemiology and Transmission
The common cold syndrome is caused by a variety of viruses that are clinically indistinguishable from each other, yet have distinct seasonal peaks (1). Rhinoviruses are the etiologic agent in 30% to 50% of colds, with seasonal peaks in fall and spring. Coronaviruses account for another 10% to 15% of colds, with a seasonal peak in midwinter. Influenza, parainfluenza, respiratory syncytial viruses, and adenovirus are etiologic agents for another 15% to 25%, although this group more commonly causes the typical influenza syndrome. A newly identified virus, metapneumovirus, has been found in children with upper respiratory illnesses, although its role in adult colds is not yet clear. Bacteria associated with pharyngitis can also cause some common cold symptoms.
The incidence of the common cold syndrome decreases with age. On average, adults have two to four colds per year; children have six to eight (1). Because person-to-person spread of colds occurs mainly in the home and at school, schoolchildren often introduce colds into a family (2).
Rhinoviruses may be transmitted from infected persons to others by direct contact with respiratory secretions as well by aerosolization of large or medium droplets of secretions in close quarters (1). Frequent, unconscious touching of virus-laden nasal mucosa contaminates the hands of infected individuals. Infectious material can survive on the hands or on inanimate objects, and hand-to-hand contact or touching these objects transfers the virus to susceptible individuals. Self-inoculation of the nares and conjunctiva (from which the virus is passed along the lacrimal ducts to the nasal passages) then completes the transmission of infection. Exposure to susceptible subjects across short distances of air may also transmit rhinovirus infection. Other viruses, such as adenovirus, respiratory syncytial virus (RSV), and influenza, are spread in similar manners, although adenovirus and influenza can also be spread by tiny droplet nuclei that may be carried by air currents and may cause lower as well as upper respiratory illness. Contagion of viruses that cause the common cold is low in community settings, relatively low from casual contact, but higher in the home and in close living quarters such as dormitories (3). Recirculation of air in an airplane cabin was not associated with an increased risk of URI (4).
Clinical Features
The correct diagnosis of the common cold is readily made by the patient. After an incubation period of 48 to
P.475
72 hours, the syndrome begins with mild malaise, rhinorrhea, sneezing, scratchy throat, and variable loss of taste and smell. These symptoms increase to maximal severity on the second to fourth day. Viral excretion and communicability are maximal during the period of the most severe symptoms. Fever usually is not present and rarely exceeds a 1°F (0.5°C) elevation in temperature. Cough and hoarseness may begin later, and their severity and duration are increased in cigarette smokers. Conversely, neither cigarette smoking nor exposure to cold ambient temperatures appears to increase the attack rate of colds. Colds usually last 1 week but may persist for 2 to 3 weeks in some cases.
|
TABLE 33.1 Examples of OTC Cold Medicationsa |
||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||
Nasal inflammation from viral colds may obstruct sinus drainage or the viral infection may extend to the sinuses themselves, mimicking the symptoms of acute bacterial sinusitis. The term acute rhinosinusitis is sometimes used to denote uncomplicated upper respiratory viral illnesses. Nasopharyngeal inflammation during a cold also commonly causes Eustachian tube dysfunction, mimicking symptoms of acute otitis media (1).
Identification of the causative virus by clinical observation or readily available laboratory testing is not possible, nor is it necessary for management. The primary challenge for the clinician is to identify patients with influenza, streptococcal pharyngitis, bacterial sinusitis, otitis media, or more unusual URIs (see Chapter 110), for whom specific antimicrobial treatment may be beneficial.
Treatment
Patients with typical URI syndromes may be assessed and managed appropriately by a telephone contact (5). Specific antiviral therapy for the uncomplicated common cold is not available, but symptomatic treatment is appropriate and may be provided by OTC medications in most cases (Table 33.1). Systemic analgesics, such as acetaminophen, aspirin, and nonsteroidal anti-inflammatory drugs (NSAIDs) relieve fever, headache, and myalgias without altering viral shedding or antibody response but may increase nasal secretions (possibly by decreasing the neutralizing antibody response) (6,7). Topical analgesics are contained in a variety of OTC preparations (e.g., phenol [Chloraseptic], menthol, dyclonine hydrochloride [Sucrets]) and may be recommended for amelioration of symptoms.
Steam inhalation did not alleviate nasal symptoms in controlled clinical trials (8), but sipping hot chicken soup (the only soup studied) increased the clearance of nasal mucus (9). Hoarseness, caused by inflammation and edema of the vocal cords, may be relieved by voice rest (see Chapter 111). Zinc lozenges taken every 2 hours may accelerate recovery of all symptoms, but evidence from controlled trials is not conclusive; the most commonly used preparation (zinc gluconate) may also cause nausea and bad taste, and the potential for toxicity exists with prolonged use (10,11).
Nasal congestion is best relieved by topical decongestants (e.g., phenylephrine [Neo-Synephrine] 0.25% or 0.5% every 4 to 6 hours, or oxymetazoline [Afrin] 0.05% twice daily) (12). Sprays rather than drops may be easier for patients, and administration in two steps may be most effective: an initial spray to decrease congestion in the membranes of the anterior nares, followed 5 to 10 minutes later by a second spray delivered deeper to the middle meatus. Patients should be cautioned against using decongestant drops or sprays for more than 3 to 5 days, to avoid
P.476
the rebound effect, rhinitis medicamentosa, an increase in nasal congestion that occurs when decongestant medication is discontinued. Systemic effects of topical decongestants are uncommon. Topical ipratropium spray (0.06%) relieves rhinorrhea and sneezing but may cause blood-tinged mucus and nasal dryness (13). Nasal corticosteroids have not been shown to have a beneficial effect in the treatment of the common cold (1).
The effectiveness of the oral decongestant, pseudoephedrine (Sudafed, 30 mg OTC or 60 mg by prescription; also available as Sudafed S.A., a sustained-action preparation containing 120 mg of pseudoephedrine, taken every 12 hours), has been demonstrated in some controlled trials (12). Although oral sympathomimetic drugs may raise blood pressure and heart rate, the effect of pseudoephedrine in recommended doses on these vital signs is usually slight. A meta-analysis demonstrated that immediate release formulations increased systolic blood pressure only about 2 mm Hg and heart rate 2 beats/per minute, on average, while there were no consistently significant effects with sustained release preparations (14). Patients with hypertension whose blood pressure was controlled showed similar effects. Some variability among studies and patients was observed, and caution is warranted in prescribing pseudoephedrine to patients who have poorly controlled or labile hypertension or cardiac disease or who have a history of episodic tachycardia.
Evidence for the effectiveness of antihistamines in treating the common cold has been equivocal (12,15). Older, sedating antihistamines (which are available OTC; Table 33.1) may be more effective because of their anticholinergic and drying effect (15). Combinations of an oral decongestant and an older antihistamine have been shown to reduce nasal symptoms and cough in controlled trials of cold treatments (12,16).
There is limited evidence that an expectorant (e.g., guaifenesin, contained in many combination cold and cough remedies) is effective in URIs (12). A later section in this chapter, Acute Bronchitis, discusses cough suppressants. Although many different combinations of multiple symptomatic treatments are available OTC and by prescription, it may be more effective, less costly, and better tolerated to recommend or prescribe individual medications targeted at a patient's principal or most distressing symptom.
Alternative and complementary treatments for the common cold are popular (see Chapter 5). There does not appear to be any prophylactic benefit of vitamin C (ascorbic acid) in preventing common colds (1,17). Numerous controlled studies of vitamin C for treating the acute symptoms of the common cold have provided conflicting results. Several earlier reviews concluded that there is no significant effect (18,19), but subsequent systematic reviews suggested a modest reduction in the duration and severity of symptoms with higher doses (e.g., 1 g or more daily) (17,20). However, some experts caution that such large doses may have adverse consequences in some people (21). Although some studies of Echinacea, a plant extract, suggested that it is beneficial in preventing or treating the common cold, recent controlled trials have had negative results and an updated systematic review reported that there is not enough evidence to recommend a specific Echinaceapreparation (22).
Antibiotics should not be prescribed for the uncomplicated cold. Randomized, placebo-controlled trials of treatment for uncomplicated, nonspecific URIs in adults have shown no benefit from antibiotics (23). Colored nasal discharge does not signal a bacterial infection or a complication of a cold; it simply indicates the presence in the discharge of polymorphonuclear leukocytes, which migrate into nasal secretions in response to cytokines induced by viral infection (23,24).
Patient Education
Because transmission of colds occurs chiefly by physical contact, it is reasonable to counsel patients and those around them that transmission can be minimized by hand washing, reduced finger-to-nose-and-eye contact, and reduced exposure to the cold sufferer. Clinicians should be particularly vigilant to avoid contact with the patient's secretions and should wash their hands carefully after examining the infected patient. Although clinicians with common colds may examine patients if they wash their hands and avoid sneezing or coughing near the patient, those with the flu syndrome should avoid patient contact.
Viral URIs may be complicated occasionally by superimposed bacterial sinusitis, otitis media, or pneumonitis. Therefore, patients should be advised to notify their health care provider of worsening symptoms suggesting one of these syndromes, each of which may require antimicrobial treatment.
Prevention
Vitamin C and Echinacea have not been shown, conclusively, to prevent the common cold. Vaccine development is complicated by the great antigenic diversity of respiratory viruses, including more than 100 serotypes of rhinoviruses and 47 serotypes of adenoviruses.
Sinusitis
The paranasal sinuses include the frontal, ethmoid, and maxillary sinuses (Fig. 33.1). These air-filled bony cavities produce up to 2 pints of mucus every day, which normally drain into the nasopharynx. The mucus is propelled by cilia that line the respiratory epithelium of the sinuses through the ostia (openings) of the sinuses and the middle meatus (passageway) in each nostril. The ostiomeatal complex, the
P.477
confluence of the drainage of the frontal, ethmoid, and maxillary sinuses, is located between the middle and inferior turbinates on each side of the nasal cavity (25) (Fig. 33.2). Symptoms of nasal and sinus congestion, pain, and discharge result from inflammation of the sinuses and obstruction of the ostiomeatal complex.
|
FIGURE 33.1. Location of the paranasal sinuses. From Williams JW, Simel DL. Does this patient have sinusitis? Diagnosing acute sinusitis by history and physical examination. JAMA 1993;270:1242. |
Acute bacterial sinusitis is an infection of one or more paranasal sinuses that occurs when the normal sinus drainage is impaired. Its symptoms overlap with those of nonbacterial and noninfectious causes of nasal and sinus inflammation, such as viral URIs, allergic and perennial rhinitis (see Chapter 30), nasal or sinus polyps, foreign bodies, local irritation and complications from swimming and diving, immune deficiency, and anatomic abnormalities that obstruct sinus drainage. Since acute bacterial sinusitis may develop as a complication of viral rhino-sinusitis, the term acute bacterial rhinosinusitis has been proposed to describe the clinical syndrome of persistent or worsening nasal and sinus symptoms following the onset of an upper respiratory illness. Some cases of acute sinusitis are an extension of a dental abscess. Nursing home or homebound patients with nasogastric tubes occasionally develop sinusitis, which may manifest as a cause of persistent fever.
|
FIGURE 33.2. Examination of the nose (with the use of a nasal speculum). From Williams JW, Simel DL. Does this patient have sinusitis? Diagnosing acute sinusitis by history and physical examination. JAMA 1993;270:1242. |
Most cases of sinusitis in healthy patients involve the maxillary and ethmoid sinuses, and the evidence for evaluation and treatment of sinusitis in outpatients is based mostly on studies of patients with maxillary sinusitis. Frontal sinusitis and infections of the deeper sinuses (e.g., the sphenoid sinuses) are more serious illnesses that usually require hospitalization (see Followup and Complications section).
Sinusitis is one of the most common diagnoses in ambulatory practice, and patients often report a previous history of sinusitis or sinus disease. Because definitive tests for the diagnosis of sinusitis, such as sinus aspiration through direct puncture or endoscopic drainage, are almost never employed in primary care, clinicians should be wary of assigning a past or current diagnosis of acute bacterial sinusitis. The main task in assessing sinus symptoms in ambulatory patients is to determine whether the patient has a sufficient likelihood of bacterial infection to warrant antibiotic therapy.
Epidemiology
It is estimated that only 2% or less of viral URIs are complicated by the development of bacterial sinusitis and that fewer than 15% of patients who seek medical attention for acute upper respiratory tract symptoms will benefit from antibiotic treatment for sinusitis (26).
Research studies have shown that the most common bacterial causes of acute, community-acquired sinusitis in adults are the common respiratory pathogens Streptococcus pneumoniae and Haemophilus influenzae (together accounting for more than half of all cases). Less commonly implicated are α-streptococci, Moraxella catarrhalis, anaerobic bacteria, Staphylococcus aureus, Streptococcus pyogenes, and gram-negative bacteria (27). Fungi occasionally cause acute sinusitis and should be considered along with cytomegalovirus and atypical mycobacteria in immunocompromised patients.
Clinical Features and Diagnosis
Patients often seek care for suspected sinusitis because of facial pain or purulent nasal discharge. When acute bacterial sinusitis is present, the pain is caused by periosteal reaction secondary to purulent inflammation behind an obstructed ostium. The pain is dull in the early stages but becomes throbbing in later stages. Coughing, dependency,
P.478
and percussion over the involved sinus may exacerbate the pain. Percussion of the teeth may be painful in maxillary sinusitis. Nontender edema of the eyelids, seen predominantly in children, may occur with uncomplicated ethmoid and maxillary sinusitis. However, the facial pain and nasal discharge associated with viral or noninfectious causes of nasal congestion often resemble the symptoms of acute bacterial sinusitis, especially when symptoms have been present less than 1 week. Other causes of facial pain to be distinguished from sinusitis are dental abscess (see Chapter 112), migraine and cluster headache, (see Chapter 87) and trigeminal neuralgia (see Chapter 87).
Examination
The initial assessment of patients with sinus symptoms should include examination of the pharynx, nose, ears, and teeth. The nostrils may be visualized more effectively with a nasal speculum, with care taken to avoid contact with the nasal septum, which is sensitive (28). Particular attention should be directed to the area of the middle meatus, between the inferior and middle turbinates, to determine whether pus is present in this area (Fig. 33.2). Tapping on maxillary teeth with a probe or tongue blade may reveal tenderness, which suggests a dental abscess. Abnormal transillumination of the maxillary sinuses was found to be useful in predicting acute bacterial sinusitis in an otolaryngology practice but was only weakly predictive in a study of primary care patients (28). If transillumination is attempted, only a concentrated halogen light source should be used; it should be placed tightly over the infraorbital rim of the patient in a completely dark room, with light transmission through the hard palate observed in the patient's open mouth. Only complete absence of light transmission, suggesting sinus opacification and infection, is helpful. Transillumination of the frontal sinuses is not useful, because they are often asymmetric in healthy patients.
Diagnostic Approach
There is no single clinical feature or easily performed diagnostic test that conclusively establishes the presence or absence of acute bacterial sinusitis. Several studies of patients with acute nasal and maxillary symptoms compared clinical findings with the results of sinus puncture and aspiration or sinus imaging studies (26). Aspiration of mucopurulent or purulent secretions, or abnormal sinus radiographs or sonograms suggestive of bacterial infection, were associated in these studies with the following clinical features: purulent nasal discharge reported by the patient or observed on examination; maxillary tooth pain; facial pain or tenderness on examination in the maxillary area; and worsening of nasal and sinus symptoms after initial improvement (Table 33.2). The likelihood of acute bacterial sinus infection increases with the number of these indicators present in a patient. In general, acute bacterial sinusitis is likely in patients who have purulent nasal discharge and unilateral maxillary tooth pain or facial discomfort for longer than 1 week. Acute bacterial sinusitis may also be present when the nasal symptoms of an upper respiratory illness have persisted for more than 10 days or worsened after 5 to 7 days (29). Patients who do not meet these criteria probably do not have bacterial infection, although severe facial pain in a patient who is febrile or toxic usually warrants empiric antibiotic treatment and further evaluation. Cultures of nasal secretions or nasopharyngeal swabs are usually contaminated with normal flora and are of no use in the evaluation.
|
TABLE 33.2 Clinical Features in Patients with Sinus Symptoms That Predict Acute Bacterial Sinusitisa |
||
|
Radiologic Examination
Diagnostic imaging of the sinuses is not recommended for most patients who present with acute symptoms suggesting sinusitis. Sinus radiographs demonstrating air–fluid levels or complete opacification are only 80% to 85% specific for acute bacterial sinusitis, although normal radiographs rule it out in 90% of cases (26,30). A single Waters (occipitomental) view is probably as accurate as a “sinus series” of multiple views (31). Many patients with acute nasal and sinus symptoms have intermediate findings on radiography (e.g., mucosal thickening), which are neither sensitive nor specific and which may be seen in patients with viral URIs (26). Computed tomographic (CT) scans have better sensitivity for detecting mucosal abnormalities, are able to visualize the ethmoid and frontal sinuses well, and are increasingly ordered in place of plain radiographs. However, the specificity of CT scans for diagnosing acute maxillary sinus infection may be poor; for example, a study of healthy volunteers experimentally infected with rhinovirus found that 87% had maxillary sinus abnormalities on CT scans (32). Diagnostic imaging of the sinuses is most helpful in the evaluation of unexplained headache and for patients who do not respond to therapy or who are toxic and require accurate diagnosis early.
Treatment
Controlled trials of antibiotic treatment for patients with suspected acute bacterial sinusitis have provided mixed results, with several studies showing no benefit of treatment (29). Systematic reviews suggest that antibiotic treatment is of moderate benefit (33), and a consensus panel reported
P.479
that 81% of antibiotic-treated patients, compared with 66% of untreated patients, were improved at 10 to 14 days (26). The apparently favorable natural history of acute bacterial sinusitis and the limited benefit of antibiotic treatment reflect the imprecision of diagnosis on clinical or radiographic grounds, and the unavoidable inclusion in controlled studies of some patients who did not really have bacterial infections and were destined to improve regardless of treatment allocation. Consensus guidelines recommend that patients with mild nasal and sinus complaints should receive only symptomatic treatment initially, and that those with worsening, persistent, or moderately severe symptoms suggesting sinusitis (see Diagnostic Approach) should be treated with antibiotics (25). This recommendation is supported by a decision analysis that found that antibiotic treatment of patients with mild to moderate symptoms was cost-effective in reducing days sick with sinusitis (34).
Choice and Duration of Antibiotics
Choosing an antibiotic to treat suspected bacterial sinusitis has become more challenging with the emergence of more widespread antibiotic resistance among the bacteria that cause it (see later section on Antibiotic Resistance) (27,35). An updated meta-analysis of clinical trials showed no advantage of newer or more broad-spectrum antibiotics compared to penicillin or amoxicillin (33), and a large retrospective analysis of almost 30,000 patients treated for sinusitis found that amoxicillin and trimethoprim/sulfamethoxazole were as effective as cephalosporins, fluoroquinolones, and the newer macrolides (36). However, all of the patients in these studies were treated before 2000, when antibiotic resistance was less of a problem. In choosing an antibiotic, clinicians should consider the severity of symptoms, the moderate benefits of treatment for most patients, cost, and convenience of dosing, as well as the possibility of antibiotic resistance. Although the rates of resistance to various antibiotics vary depending on locality, this information may not be available to clinicians. Frequent contact with children in day care, or recent or frequent use of antibiotics, may be used as proxies for an increased risk of antibiotic resistance in individual patients. For many patients, amoxicillin 500 mg three times daily, trimethoprim 160 mg/sulfamethoxazole 800 mg (i.e., the double-strength dosage) twice daily, or doxycycline 100 mg twice daily are appropriate choices for initial therapy. For patients with moderately severe symptoms, those who are toxic, and those in whom antibiotic-resistant bacteria may be present amoxicillin/clavulanate 500 mg/125 mg three times daily or 875 mg/125 mg twice daily, cefuroxime axetil 500 mg or cefpodoxime 200 mg twice daily, or a fluoroquinolone once daily should be prescribed (see Table 33.3).
The recommended duration of therapy for uncomplicated acute bacterial sinusitis is 10 days. Although one randomized trial compared a 3-day to a 10-day course of trimethoprim/sulfamethoxazole and found no difference in outcome at 14 days, it lacked sufficient power to detect a difference in the rates of relapse, recurrence, or serious complications (37). Azithromycin is approved by the Food and Drug Administration (FDA) for shorter courses of treatment for sinusitis (Tri-pak 500 mg once daily for 3 days, and Zmax extended-release suspension 2 g once only), but its use may be most appropriate for patients with mild symptoms who are allergic to penicillins or for whom adherence to a longer regimen may be problematic. Shorter courses of a high-dose formulation of amoxicillin/clavulanate (Augmentin XR) or of a new ketolide antibiotic, telithromycin (Ketek), may be effective (see Table 33.3 for dosing), but outcome data are limited.
Other Treatments
Although there is no conclusive evidence to support the use of adjunctive therapies for acute sinusitis, improvement in sinus drainage may be helpful. A topical decongestant, such as phenylephrine (Neo-Synephrine) 0.25% or 0.5% every 4 to 6 hours, or oxymetazoline (Afrin) 0.05% twice daily, may be recommended (see section on Common Cold above for method of administration). Use of topical decongestants should be limited to approximately 5 days to avoid rebound congestion. An oral (systemic) decongestant, such as pseudoephedrine (30 to 60 mg every 6 to 8 hours), may have a more reliable therapeutic effect in a congested nose and may be continued for the duration of symptoms (Table 33.1; also see Common Cold above for precautions). The effectiveness of a nasal corticosteroid in improving the resolution of sinusitis is uncertain, but it may be helpful in patients with a history of allergic or chronic nasal symptoms (29).
Pain relief is important, and a short course of an opiate analgesic may be required in addition to OTC analgesics and NSAIDs. Patients who plan to fly, especially in nonpressurized aircraft, should take an oral decongestant before takeoff, supplemented with topical decongestant spray every 4 hours.
Followup and Complications
Resolution of facial pain, headache, and fever is expected within several days of starting treatment. If there is no improvement within 3 to 5 days, or if symptoms recur within 2 weeks, then antibiotics should be changed to a broad-spectrum agent from a different class (e.g. amoxicillin/clavulanate or a fluoroquinolone) (27,35). When there are worsening or persistently severe symptoms, the diagnosis should be reassessed by radiography or computed tomography (CT) scan, and referral to an otolaryngologist for antral puncture or endoscopic drainage may be advisable. For toxic patients and those with suspected frontal or ethmoid sinusitis, imaging, consultation with an
P.480
otolaryngologist, and hospitalization for drainage, culture, and definitive parenteral antibiotics should be considered at initial presentation. Patients with severe facial pain benefit from early antral puncture for pain relief.
|
TABLE 33.3 Oral Antibiotics Used for Ambulatory Treatment of Common Respiratory Infections |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Serious complications of acute bacterial sinusitis are unusual and have been estimated to occur at a rate of only 1 in 10,000 cases (34). They constitute medical emergencies because they represent direct extension of infection to adjacent orbits, bone, blood vessels, and the central nervous system. Nontender periorbital edema indicates restriction of orbital venous outflow through congested ethmoid veins; it is not associated with decreased visual acuity and is appropriately managed with vigorous medical therapy. However, tender periorbital swelling associated with proptosis and chemosis represents orbital cellulitis and requires immediate referral to an otolaryngologist. Subsequent progression of cellulitis to subperiosteal or orbital abscess, associated with ophthalmoplegia and loss of vision, requires emergency surgical drainage. Osteomyelitis is most often a complication of frontal sinusitis. Cavernous sinus thrombosis should be suspected in the patient with signs of orbital complications plus extreme toxicity. Intracranial extension is rare but life-threatening, manifesting most commonly as meningitis. Abscesses in the brain and epidural and subdural spaces manifest more insidiously. Frontal lobe abscess may manifest as mild headache, low-grade fever, malaise, and personality change. In poorly controlled diabetics and immunocompromised hosts, invasive fungal infections of the nose and sinus can be severe and deadly; rhinocerebral mucormycosis may be recognized by a black eschar on the nasal turbinates.
P.481
Chronic Sinusitis
When the symptoms of acute sinusitis, especially pain and fever, subside with therapy, purulent nasal discharge may continue. Despite persistence of radiologic changes, this stage usually resolves after an additional 2 to 3 weeks of conservative management with oral decongestants.
Sinusitis is classified as subacute when symptoms are prolonged for 4 to 12 weeks, and chronic when symptoms persist longer than 12 weeks. Chronic sinusitis results from obstruction to sinus drainage and loss of the normal ciliated epithelial lining of the sinus cavity, sometimes leading to a low-grade infection by anaerobic and aerobic bacteria. Acute exacerbations may occur, caused primarily by the organisms that most commonly cause acute sinusitis (H. influenzae and S. pneumoniae). Chronic sinusitis may complicate asthma associated with allergic rhinitis, and may be a manifestation of systemic diseases such as sarcoidosis, Wegener granulomatosis, immunoglobulin A deficiency, and human immunodeficiency virus (HIV) infection.
Diagnosis and Management
Persistent purulent nasal discharge and postnasal drip with associated cough, despite adequate medical therapy, are the primary features of chronic sinusitis. Facial pain and tenderness may be minimal or absent. A CT scan may confirm the presence of sinus mucosal inflammation and thickening and indicate whether specific anatomic abnormalities such as polyps, mucoceles, or tumors may be involved.
Amoxicillin for 1 month may be used if the patient has not been treated with antibiotics recently; when chronic symptoms develop after an initial course of antibiotic treatment, or there are other reasons to suspect antibiotic resistance (see Choice and Duration of Antibioticssection), amoxicillin/clavulanate or clindamycin are alternatives (Table 33.3). Continued use of systemic decongestants, as well as a trial oftopical corticosteroids (in dosages similar to those used for allergic rhinitis; see Chapter 30), may facilitate resolution of symptoms in chronic sinusitis. For patients whose symptoms do not resolve with one or more of these empiric treatments, referral to an otolaryngologist for consideration of endoscopic or surgical drainage is recommended.
Flu Syndrome
“Flu” is said to have been named by two Italian astrologers, who believed it was caused by the influence, or “influenza,’’ of the stars. Although the term “flu-like illness” is commonly used by patients and clinicians to describe any malady thought to result from a virus, including gastroenteritis, this section describes only illnesses caused by respiratory pathogens. Flu syndrome presents as the abrupt onset of malaise, myalgia, headache, and fever. Coryza and sore throat are also present. Illness is severe for 3 to 14 days, and convalescence lasts for 1 to 4 weeks. The majority of cases of flu syndrome are caused by the influenza virus during annual winter epidemics. Other viruses, especially parainfluenza and respiratory syncytial viruses and adenovirus, produce the same clinical syndrome and may coinfect patients with influenza (38,39) (Fig. 33.3).
Epidemiology
There are two genera, or types, of influenza virus that infect humans: influenza A and influenza B. Influenza A is further subtyped into strains according to variations in the surface glycoproteins, hemagglutinin (H1 through H15) and neuraminidase (N1 through N9) (40). Epidemic spread of the influenza virus is caused by the appearance in nonimmune populations of new antigenic variations of the virus. Variations of the hemagglutinin and neuraminidase antigens occur almost annually in influenza A, and less often in influenza B. Major variation is called antigenic shift and results in pandemic spread of a new strain, almost always type A, throughout regions of the world where there is little natural immunity. The most recent pandemics occurred in the winters of 1957–1958, 1968–1969, and 1977–1978, and they varied considerably in severity. Between pandemics, minor antigenic variation, termed antigenic drift, occurs frequently, resulting in a new strain and epidemics during the winter almost every year. Such interpandemic spread, although less dramatic, accounts for greater cumulative morbidity and mortality because of its annual occurrence. In recent years, epidemic influenza has occurred regularly and has had a major influence on mortality and hospitalizations in this country, especially among the elderly and individuals with chronic illness. It is estimated that influenza is responsible for approximately 36,000 deaths and more than 100,000 excess hospital admissions yearly (41,42). Although it is often impossible clinically to distinguish infections caused by type A or type B, influenza A is responsible for greater excess mortality than type B.
Influenza virus typically circulates in the Northern Hemisphere from late November through March (Fig. 33.3). During this time there is usually a more intense peak of illness, attributable to a predominant strain, which lasts for several weeks. There may also be a second, less intense peak caused by another strain, which sometimes heralds the predominant subtype the following year. Influenza also occurs sporadically at other times of the year, and it may affect travelers to other parts of the world where influenza is circulating, as well as individuals exposed to persons from those areas.
Influenza virus is highly contagious and is spread in both community and home settings (3). It is transmitted by direct contact with infected secretions as well by large- or medium-droplet aerosols dispersed by sneezing, coughing, or talking. It may also be spread via tiny droplet
P.482
nuclei, which are disseminated by air circulation systems in closed spaces such as airplane cabins. The incubation period is 18 to 72 hours. Viral shedding persists for 5 to 10 days, but virus is present in high titer in secretions for only 48 hours after the onset of clinical illness. In the community, person-to-person transmission is rapid, with spread initially among children, then adults. In local epidemics, the incidence of cases reaches a peak in 2 to 3 weeks and persists for only 5 to 6 weeks.
|
FIGURE 33.3. Schematic representation of the occurrence, etiology, and mortality from febrile respiratory illnesses in relation to time of year. Data are from Houston, Texas, 1975–1981. From Glezen WP. Serious morbidity and mortality associated with influenza epidemics. Epidemiol Rev 1982;4:25. |
Beginning in 2003, the widespread occurrence of a novel influenza type, human avian influenza (H5N1), was found in poultry in several countries in southeast Asia, and then subsequently in other areas in Asia and Europe. Within several years, over 100 cases of H5N1 were documented in persons exposed to birds, but only several instances of human-to-human transmission have occurred (43). Despite the apparently sporadic nature of human infection with this and other strains of avian influenza, because of concerns that further viral genetic variation and adaption to human infection might lead to a global pandemic, governmental health authorities have instituted more intensive influenza surveillance.
Clinical Features and Diagnosis
Uncomplicated influenza, type A or B, has an abrupt onset of systemic symptoms including fever, chills, headache, myalgias, and malaise. The fever, which may rise to 106°F
P.483
(41°C) in some cases, typically lasts 3 days, although it may persist for 5 to 7 days. Headache and myalgias involving the back, arms, legs, and occasionally the eyes are the predominant symptoms, persisting as long as the fever. Respiratory symptoms, such as nonproductive cough, nasal discharge, hoarseness, and sore throat, appear as systemic symptoms wane. Cough and weakness usually subside after 2 weeks but may persist longer. In the elderly, myalgias and sore throat are less common, whereas dyspnea is a more common symptom (40).
Physical findings include general toxicity, flushed face, hot skin, watery red eyes, clear nasal discharge, tender cervical lymph nodes, and occasionally localized rales in the chest. The white cell count and differential, indicated only if the patient is toxic or pneumonia is suspected, usually demonstrate mild neutropenia and relative lymphocytosis caused by absolute granulocytopenia.
Because specific antimicrobial therapy for influenza infections is available, it is important to determine whether flu syndrome is likely to be caused by this virus. The best clinical predictors of influenza virus as the cause of flu syndrome are fever greater than 100°F (37.8°C), cough, and abrupt onset (44). However, during influenza epidemics, when the prevalence of influenza infection may be 65% or higher among individuals with any respiratory symptoms, this combination is neither sensitive nor specific. Therefore, influenza should be suspected in patients who have respiratory illness during flu season (November through April), and it should be presumed to be the cause of acute respiratory illness with fever and cough when influenza is known to be circulating in the community. Clinicians can obtain information about influenza epidemics from the Centers for Disease Control and Prevention (CDC) website (http://www.cdc.gov/flu/weekly/fluactivity.htm) and Voice Information System (influenza update at 888-232-3228) and from state and local health departments.
Since influenza vaccination is likely to be only 30% to 40% effective in preventing upper respiratory influenza illness in targeted patients (although it is more effective in preventing influenza-associated pneumonia, hospitalization, or death), flu syndrome in vaccinated patients still may be caused by the influenza virus and respond to specific therapy (41,44).
Diagnostic Tests
Influenza virus infection may be diagnosed by viral culture, polymerase chain reaction (PCR) assay, or specific serology (comparing acute and convalescent titers). These tests require specialized techniques, and the results are not available until days to weeks after a patient is evaluated. Immunofluorescence antibody staining and enzyme immunoassay (EIA) tests may be performed on respiratory secretions with results available in several hours, but these tests are not readily available in ambulatory practice. A variety of rapid test kits for point-of-care analysis of nasopharyngeal swabs or nasal aspirates or washes are available to clinicians and provide results within 30 minutes (44). Most of these tests are moderately complex and require CLIA (Clinical Laboratory Improvement Amendments from the Centers for Medicare and Medicaid Services) certification. Some of the rapid diagnostic tests can distinguish between influenza A and B. Reported sensitivity is 70% or more (with no clear advantage for any specific kit), but specificity is better, usually 90% or more. Therefore, a positive test reliably confirms a diagnosis of influenza infection, but a negative test does not exclude influenza or the potential benefit of specific treatment. Testing is not likely to be helpful during an influenza epidemic, when the pretest probability of influenza infection is high in a person with typical symptoms, but it may be most useful when influenza is not known to be circulating widely or when the clinical diagnosis is uncertain (44). Information about diagnostic testing for influenza, including available rapid testing kits, may be found on the CDC website (http://www.cdc.gov/flu/professionals/labdiagnosis.htm).
Treatment
Options for the specific antimicrobial treatment of influenza include the older, adamantine compounds, amantadine and rimantadine, and two newer neuraminidase inhibitors, zanamivir and oseltamivir (Table 33.4). Amantadine and rimantadine are effective only against influenza A, whereas the neuraminidase inhibitors are active against both A and B strains.
All of these drugs attenuate clinical disease by reducing fever by 50% and by shortening the duration of illness by 1 or 2 days. Reductions in illness duration may be somewhat greater when influenza symptoms or fever are more severe, and there is some evidence that viral shedding (and contagion) decreases with treatment. These benefits were observed when the drug was administered within 24 to 48 hours after onset of illness, although some experts suggest treating high-risk patients or severely ill patients who present within 3 to 4 days after symptom onset (45). Evidence for the effectiveness of these drugs in treating influenza illness is based mostly on studies in younger, healthy people with uncomplicated influenza illness, and data in patients at high risk for complications, including the elderly, are limited. There is only limited evidence that neuraminidase drugs are effective in preventing serious complications of influenza (e.g., pneumonia, exacerbations of chronic diseases) (41,46).
Table 33.4 shows dosages, routes of administration, adverse effects, and costs of these drugs. Previously, drug resistance was not a clinically significant problem in treating influenza, but in 2006 the predominant influenza A strain (H3N2) showed widespread resistance to amantadine
P.484
and rimantadine. Resistance to the neuraminidase inhibitors remained rare. If amantadine or rimantadine is used to treat influenza illness, it should be discontinued after 3–5 days, or within 24–48 hours after symptoms have resolved; oseltamivir and zanamivir are given for 5 days (41).
|
TABLE 33.4 Antimicrobial Drugs for the Treatment of Influenza |
||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||
Drug treatment should be considered for patients at high risk for morbidity and mortality who develop an influenza-like illness in a community where local influenza activity has been reported. Groups at high risk include patients with chronic pulmonary, cardiovascular, metabolic, neuromuscular, or immunodeficiency diseases and patients taking immunosuppressive medications. Adults whose activities are vital to community function, including selected hospital personnel, also should be considered for drug treatment of influenza.
Choice of therapy depends on consideration of the type of influenza circulating in the community, the safety, cost, convenience and availability of the drugs, and the concerns about resistance described above. Zanamivir is delivered via an inhalation device that requires assembly, and elderly patients have been shown to have difficulty with its use (47). Some patients with obstructive lung disease have had bronchospasm after using zanamivir, so it is generally not recommended for patients with asthma or chronic obstructive pulmonary diseases (COPD).
Supportive measures are important for symptomatic relief. Rest and adequate fluid intake should be advised. NSAIDs, including aspirin 650 mg every 3 to 4 hours, or acetaminophen 650 to 1,000 mg every 4 to 6 hours (maximum 4 g daily), reduce headache, fever, and myalgia. Aspirin should be avoided in children. Relief of nasal discharge may be obtained by agents discussed previously (see Common Cold). Relief of cough with cough suppressants is discussed later (see Acute Bronchitis).
Complications
Pulmonary complications exhibit a spectrum of severity, from mild airway hyperreactivity without pulmonary infiltrates, to segmental influenza pneumonia or secondary bacterial pneumonia, to fulminant bilateral influenza pneumonia with the acute respiratory distress syndrome (ARDS). Patients should be advised that dyspnea, hemoptysis, wheezing, purulent sputum, fever persisting longer than 7 days, and, rarely, dark urine or severe muscle pain herald complications that demand prompt medical attention and sometimes hospitalization.
Airway hyper-reactivity may occur after some influenza infections and other viral respiratory tract infections manifested clinically by bronchospasm, coughing, or both (48,49). Patients with asthma or chronic bronchitis have even greater bronchoconstrictor responses to influenza infection because of their underlying bronchial smooth muscle hyperreactivity. Airway hyper-reactivity may be demonstrated for 3 to 8 weeks after infection by influenza or other viruses, and occasionally it may last for 4 to 6 months, even in patients who are not atopic. Both cough and wheezing after an otherwise uncomplicated flu syndrome may be treated with a trial of an inhaled bronchodilator (seeChapter 60) as needed and at bedtime. Patients troubled particularly by nighttime cough may obtain additional relief with 15 to 30 mg of codeine at bedtime.
During influenza epidemics, there is an increase in the incidence of pneumonia, although influenza often is not recognized by clinicians as a primary or contributing
P.485
cause in patients who are admitted to the hospital with community-acquired pneumonia (CAP) (50). The incidence of bronchitis and pneumonia associated with influenza varies with age: it is low in patients younger than 50 years of age and very high in patients older than 70 years of age. The characteristics of pneumonias associated with influenza are described in the section on Lower Respiratory Tract Infections below.
Nonpulmonary complications of influenza are unusual. Myositis, with thigh pain and inability to walk, occurs occasionally in children and adolescents. Severe myositis with myoglobinuria and acute renal failure has been observed in adults after both influenza A and B. Guillain-Barré syndrome, encephalitis, and transverse myelitis are neurologic complications associated rarely with influenza infection, but no firm causal relationship has been established. Reye syndrome (encephalopathy and fatty liver) is a rare but severe complication of influenza, usually type B; patients present with a change in mental status and progress to coma and hepatic failure. The mean age at attack is 6 years, and the incidence has fallen markedly in recent years. The syndrome is rare in adults and, unlike the situation in children, is not associated with aspirin. With the exception of mild myositis, all of the nonpulmonary complications of influenza require hospitalization for differential diagnosis and management.
Prevention
Chapter 1 discusses the use of influenza vaccine and drug prophylaxis in ambulatory practice.
Pharyngitis
Sore throat is among the most common symptoms seen in ambulatory medical practice. Most acute episodes of pharyngitis in adults are self-limited and of short duration, and significant complications are rare. The most important task in the evaluation of patients who complain of sore throat is to identify group A streptococcal and other bacterial infections, for which antibiotic treatment is appropriate, and to recognize less common causes of pharyngitis associated with more serious illnesses. Although many adult patients with sore throat are treated with antibiotics (51), it is increasingly recognized that this practice is often not appropriate.
Epidemiology
Pharyngitis in adults is caused by a variety of viral and bacterial pathogens, with no single etiology predominating (52,53). The majority of cases are caused by common viruses, most often rhinovirus, coronavirus, and adenovirus. Streptococcal bacteria, predominantly group A β-hemolytic Streptococcus (GABHS) species, account for only 15% of cases or less. Mycoplasma pneumoniae, the TWAR strain of Chlamydia pneumoniae, Arcanobacterium (formerly Corynebacterium) haemolyticum, Neisseria gonorrhoeae, H. influenzae type b, Corynebacterium diphtheriae, Candida species, respiratory syncytial virus, influenza types A and B, parainfluenza, herpes simplex virus, adenovirus, and Epstein-Barr virus cause pharyngitis infrequently.
Group A β-Hemolytic Streptococcal Pharyngitis
Clinical Features
Only 5% to 15% of episodes of acute pharyngitis in adults are caused by GABHS (54). It occurs most commonly in the winter and spring. Individuals who have regular contact with children (e.g., parents, teachers), or who have been exposed to others with diagnosed streptococcal pharyngitis, are more likely to have GABHS. The incubation period is 2 to 4 days, followed by the abrupt onset of sore throat, malaise, fever, and headache. Mild neck stiffness and gastrointestinal symptoms are sometimes present (55). All of the features of the classic syndrome, including fever, tender anterior cervical and tonsillar lymph nodes (at the angle of the jaw), and enlarged tonsils with white exudate, occur together infrequently, and each of these features may occur in other types of pharyngitis. Importantly, cough, hoarseness, and rhinorrhea are not usually present in the patient with strep throat. A distinctive scarlatiniform rash (scarlet fever)occasionally occurs. It is characterized by a diffuse red blush appearing on the trunk early in the disease, spreading centrifugally, blanching with pressure, and acquiring a sandpaper texture; 1 week later the skin desquamates, particularly over the palms and soles. This rash is not seen in most patients with GABHS pharyngitis. A similar rash also occurs in toxic shock syndrome and Kawasaki syndrome and a rash associated with A. haemolyticum pharyngitis is localized to the trunk and does not desquamate.
Diagnosis
Because individual clinical findings are nonspecific, the diagnosis of streptococcal pharyngitis relies on clinical prediction rules, rapid antigen tests, or throat culture. Several clinical prediction rules for GABHS pharyngitis have been developed (55); the Centor criteria (56) are simple and straightforward, have been validated prospectively, and have emerged as a consensus tool (54,55) (Table 33.5). These criteria include: tonsillar exudates, tender anterior cervical lymphadenopathy, history of fever (temperature greater than 100.4°F [38°C]), andabsence of cough. The presence of three or four of these criteria has a sensitivity of 75% and a specificity of 75% for GABHS pharyngitis, using throat culture as the reference standard. For most
P.486
adults, this results in a positive predictive value of 40% to 60% when three or four criteria are present, and a negative predictive value of approximately 80% if none or only one of the criteria is present (54).
|
TABLE 33.5 Clinical Features in Patients with Sore Throat That Predict Group A β-Hemolytic Streptococcal (GABHS) Pharyngitisa |
||
|
Rapid antigen tests, which employ enzyme and optical immunoassay methods to detect GABHS carbohydrate products, are commercially available for point-of-care use in ambulatory practice. The specificity of these tests is reported to be greater than 95%, although the sensitivity is lower, e.g., 80% to 90% (53). For the average prevalence of GABHS in adults with sore throat (i.e., 10%), rapid antigen tests have a positive predictive value of approximately 65% and a negative predictive value of 98%. Therefore, the use of these tests alone (that is, without regard to clinical features) does not reliably establish the diagnosis of strep throat, although they are useful in conjunction with clinical criteria. A negative rapid strep test does accurately confirm the absence of GABHS.
Throat culture has been the gold standard for diagnosing GABHS, although a significant disadvantage is that results are not available for 24 to 48 hours. A small percentage of false negative tests may be caused by inadequate specimen collection or improper handling. False positive tests may occur if the patient is an asymptomatic carrier of GABHS and the acute pharyngitis is caused by another pathogen; this is estimated to occur in only 2% to 4% of adolescents and adults (55). Currently, many physicians prefer to send throat swabs in transport media to commercial laboratories. Inexpensive office throat culture kits are also available and have a high sensitivity (approximately 95%).
The accuracy of rapid antigen tests and throat cultures depends on proper collection of the throat swab. The pharynx must be viewed adequately, with elevation of the soft palate and depression of the posterior tongue. Use of a tongue blade and the classic “ahh” phonation by the patient may help; sometimes not having the patient stick out the tongue, or having the patient pant, is useful. The tonsillar tissue and posterior pharynx should be swabbed vigorously; adequate collection often induces the gag reflex.
Diagnostic Approach
Adult patients with sore throat should be screened for the four clinical findings (Centor criteria), as outlined previously. The clinician should also be alert to aspects of the history, symptoms, and signs that suggest other, potentially treatable or serious causes of pharyngitis (seeOther Bacterial Causes of Pharyngitis and Noteworthy Viral Causes of Pharyngitis sections). Patients with none or only one of the four clinical criteria should not receive further testing or antibiotic treatment, because they are unlikely to have GABHS. Patients with two or more of the criteria should be tested with a rapid antigen kit or throat culture and treated with antibiotics if the result is positive. One consensus guideline suggests that patients with three or four of the criteria may be treated with antibiotics empirically (54); however up to one half of such patients may not have GABHS and will be treated unnecessarily. Symptomatic family contacts of patients with streptococcal pharyngitis should be tested and should be treated with antibiotics if the tests or cultures are positive. Routine testing of asymptomaticfamily members is not indicated.
Acute Rheumatic Fever and the Rationale for Antibiotic Treatment
The benefits of treating sore throat with antibiotics, especially if GABHS can be confirmed, include prevention of acute rheumatic fever and suppurative complications and perhaps more prompt relief of symptoms and interruption of contagious spread of pharyngitis. There is a growing recognition, however, that the magnitude of the benefits of antibiotic treatment may not be as large as previously believed.
Acute rheumatic fever is a clinical syndrome of nonsuppurative inflammatory lesions of the heart, joints, and central nervous system that follows GABHS pharyngitis (57,58). Diagnosis is based on the Jones criteria: two major criteria (carditis, polyarthritis, chorea, subcutaneous nodules, and erythema marginatum), or one major criterion and two minor criteria (fever, arthralgia, heart block, elevated acute-phase reactants including granulocytosis, erythrocyte sedimentation rate, and C-reactive protein). Evidence of recent streptococcal infection must be confirmed by either positive throat culture, streptococcal antigen test, or elevated or rising antistreptococcal antibodies. One third of acute rheumatic fever cases occur after asymptomatic streptococcal infection, but almost all cases are associated with a rise in serum antistreptolysin O (ASO) titers. The latent period between clinical streptococcal pharyngitis and onset of acute rheumatic fever ranges from 1 to 5 weeks, with a mean of 19 days. Clinical trials have confirmed that appropriate antibiotic treatment of GABHS pharyngitis is highly effective in preventing rheumatic fever if administered within approximately 1 week after the onset of illness.
The incidence of acute rheumatic fever declined dramatically in the United States and other industrialized countries during the last century, falling to extremely low
P.487
levels during the 1960s and 1970s. It continued to be endemic in developing countries, where it accounts for up to 40% of all cardiovascular disease. Outbreaks among school-age children and young adult military recruits in the United States in the 1980s raised fears about a resurgence of rheumatic fever, but since then its occurrence has continued to decrease to only 1 case per 1 million people per year. Although the CDC suspended reporting of acute rheumatic fever in 1995, it has remained endemic at low levels in the region surrounding Salt Lake City, Utah (59).
Because almost all of the recent, very infrequent cases of acute rheumatic fever have been in children or young adults living in close quarters, the risk in most adults after GABHS is likely to be extremely low. The reasons for the overall decline and periodic resurgence of rheumatogenic GABHS infections are not completely understood or predictable, so clinicians should be aware of emerging trends in streptococcal disease.
Suppurative complications, such as peritonsillar or retropharyngeal abscess (see Pharyngeal Abscesses section), are very rare, although a quantitative review demonstrated a further reduction in their occurrence with antibiotic treatment of GABHS pharyngitis (60). However, patients with these infrequent complications often already have them at initial evaluation, or have a negative test initially, or were treated initially with appropriate antibiotics (54). Although antibiotics are often recommended for patients with GABHS pharyngitis to prevent contagion, the utility of this approach for adults in noninstitutionalized settings is not known. There is good evidence from controlled clinical trials and systematic reviews that antibiotic treatment of suspected or confirmed GABHS provides some relief of symptoms; however, therapy must be initiated within 2 to 3 days after the onset of illness, and the benefit is limited to shortening the duration of symptoms by 1 to 2 days (54,60).
Patients in whom GABHS is clinically suspected or confirmed by testing should be treated. In three additional groups of patients who present with sore throat, antibiotic treatment for streptococcal pharyngitis should be started and throat cultures obtained: patients with a history of rheumatic fever not currently taking prophylaxis, young patients with a strong family history of rheumatic fever, and all new cases of pharyngitis in an explosive epidemic of streptococcal disease in close populations such as groups of military personnel or students living in dormitory settings. Local health authorities should be notified promptly in this last situation.
Choice of Antibiotics
The preferred therapy for GABHS pharyngitis is parenteral benzathine penicillin, 1.2 million units given once intramuscularly, because it obviates nonadherence and is the only specific treatment proven in clinical trials to prevent rheumatic fever. If oral therapy is given, the recommended regimen is penicillin V, 500 mg twice daily for 10 days (53). Resistance to penicillin has not been a problem in GAHBS. For patients who are allergic to penicillin, a 10-day course of erythromycin 250 mg every 6 hours or 500 mg twice daily or a first-generation cephalosporin, is recommended. The use of newer macrolides such as azithromycin to treat GABHS is discouraged because of their increased costs and potentiation of antibiotic resistance. Posttreatment cultures should be performed only if there is a history of rheumatic fever in the patient or in a household contact.
Symptomatic Treatment
In addition to recommending antibiotics only for patients with suspected or confirmed GABHS pharyngitis, clinicians can encourage all patients with sore throat to try antipyretics and systemic and topical analgesics in appropriate doses (see section on Common Cold), along with supportive measures such as gargling.
Other Bacterial Causes of Pharyngitis
Pharyngitis caused by A. haemolyticum is characterized by exudative pharyngitis, a scarlatiniform rash, fever, adenopathy, and a negative test for GABHS (52). It may be treated with erythromycin 500 mg orally twice daily for 10 days.
Gonococcal pharyngitis should be considered in patients who complain of sore throat in association with urethritis or vaginitis; it occurs alone without genital symptoms in fewer than 5% of cases and must be diagnosed by throat culture. Special culture techniques (including the use of Thayer-Martin medium, which is available in kits for office cultures, or specific swab kits with appropriate transport media) should be used to detect gonorrhea in specimens from patients practicing orogenital sex. Calcium alginate swabs should be used, because ordinary cotton swabs contain fatty acids inhibitory to gonococcal growth. Gram staining of a direct pharyngeal smear is insensitive and nonspecific. For throat cultures, N. gonorrhoeae must be distinguished from Neisseria meningitidis and Neisseria lactamica by carbohydrate fermentation and serology; therefore, cultures should be sent to state or regional laboratories. Patients with gonococcal infections are often coinfected with chlamydia; although coincident chlamydial pharyngitis is unusual, it is recommended that patients with gonococcal pharyngitis also be treated empirically for possible genital chlamydia infection. Treatment of gonococcal infections, including the oropharynx, are discussed in detail in Chapter 37. Clinicians may also refer to the CDC website (http://www.cdc.gov/STD/treatment/, accessed 3-27-06) for updates.
Diphtheria is exceedingly rare in this country. Pharyngitis and skin infections caused by C. diphtheriae have occurred recently only in persons from disadvantaged groups
P.488
who were inadequately immunized. It should be suspected when there is a grayish membrane in the anterior nares or on the tonsils, uvula, or pharynx. Treatment must begin before bacteriologic confirmation and requires hospitalization for strict isolation, bed rest, close observation, diphtheria antitoxin, and erythromycin or penicillin for 14 days. Chapter 18 discusses vaccination against diphtheria and management of exposed contacts.
Throat cultures sometimes grow other bacteria, such as pneumococci, staphylococci, group B, C, or G streptococci, and various gram-negative enterobacteria. These species colonize the pharynx and rarely cause pharyngitis, and patients who harbor them generally should not be treated with antibiotics. Some outbreaks of pharyngitis have been traced to streptococci of groups C and G, and patients with persistent sore throat and throat cultures positive for these organisms may be treated with shorter courses of the antibiotics listed for GABHS (54).
Vincent angina is an anaerobic bacterial infection of the pharynx characterized by fever, tender lymphadenitis, a large grayish-brown pseudomembrane in the pharynx, and very foul odor. It is a complication of acute necrotizing ulcerative gingivitis (see Chapter 112). Hospitalization for antimicrobial treatment with penicillin or tetracycline is the appropriate management plan.
Noteworthy Viral Causes of Pharyngitis
Infectious mononucleosis (see Chapter 58) is characterized by the clinical triad of sore throat, fever, and lymphadenopathy. It can be distinguished on clinical grounds from streptococcal infection only when hepatosplenomegaly and a maculopapular skin rash (similar to a drug eruption or rubella, typically precipitated by ampicillin) are present. Palatal petechiae may be seen in mononucleosis but may also occur with rubella or streptococcal pharyngitis. Pharyngoconjunctival fever, caused by several adenovirus strains, is usually accompanied by influenza-like symptoms and can be distinguished by concurrent conjunctivitis in one third of cases and a history of swimming pool exposure 1 week before onset. It has also been reported among military recruits (52). Oropharyngeal infection with herpes simplex virus or coxsackie A virus is distinguished by the presence of mucosal vesicles or ulcers (see Chapter 112). The vesicular enanthem in the pharynx caused by coxsackie viruses is sometimes called herpan-gina. The acute retroviral syndrome caused by HIV infection may manifest with fever and nonexudative pharyngitis; systemic symptoms and occasionally a rash occur also (see Chapter 39).
Chronic or Relapsing Sore Throat
Some patients describe a sore throat of several weeks’ duration at their first visit. Others have either a prolonged course after an illness that began as a typical acute pharyngitis syndrome or frequent recurrence of sore throats. Table 33.6 lists the conditions that can cause prolonged or recurrent pharyngitis. (Most are discussed in more detail elsewhere in this book, as indicated in the table.)
|
TABLE 33.6 Causes of Chronic or Relapsing Sore Throat |
||||||||||||||||||||||||||||||||||||||||||
|
Chronic tonsillitis or recurrent pharyngitis is a clinical diagnosis made in patients with frequent sore throats (more than six in 1 year, or three or more episodes in 2 or more years), very large tonsils, and chronically enlarged, periodically tender lymph nodes. The effectiveness of tonsillectomy to alleviate chronic tonsillitis in children has been controversial (61); there have been no controlled trials in adults, and its indications in adults are not well established. Postoperative pain and hemorrhage are the most common complications, the latter occurring in 0.5% to 2% of patients (but perhaps more often in adults) (62).
β-Lactamase–producing organisms in the pharynx, including S. aureus, Haemophilus species, Bacteroides species, and Branhamella catarrhalis, can inactivate penicillin and protect mucosal streptococci; these conditions may underlie some cases of chronic or recurrent pharyngitis. Patients who have multiple episodes of GABHS pharyngitis should receive clindamycin 300 mg twice daily or amoxicillin/clavulanate 500 mg twice daily (53).
Pharyngeal Abscesses
Occasionally, after several days of symptoms of a URI, a patient may develop a complicating infection of one of the closed compartments adjacent to the pharynx. The most common of these pharyngeal abscesses are peritonsillar abscess (also known as “quinsy”) andretropharyngeal abscess. If one of these conditions is suspected, the patient should
P.489
be referred immediately for evaluation and management by an otolaryngologist. Table 33.7 lists other conditions to consider when the patient is unable to swallow saliva because of pain.
|
TABLE 33.7 Differential Diagnosis of Severe Throat Pain, Odynophagia, and Inability to Swallow Saliva |
||||||||||||
|
Patients with peritonsillar abscess develop severe odynophagia; they are unable to take liquids and also may be unable to swallow their own saliva, resulting in early dehydration. The voice acquires a muffled quality, and trismus may be present. Fever, malaise, and systemic toxicity are typical. Dramatic relief may occur if the abscess drains spontaneously before the patient seeks medical attention. On physical examination, there is a swelling of the anterior tonsillar pillar at its superior pole. The involved tonsil itself may or may not be enlarged, but it is displaced medially. This condition is almost always unilateral. Half of the cases of peritonsillar abscess are caused by group A S. pyogenes.
The symptoms of retropharyngeal abscess are similar to those of peritonsillar abscess. In addition, there may be respiratory embarrassment if the process extends inferiorly toward the larynx. Trismus is uncommon. On examination, a swelling in the posterior oropharynx is readily seen. Lateral soft tissue radiographs of the neck may disclose expansion of the soft tissue density in the posterior pharyngeal space.
Management
Unless the airway or swallowing is compromised, needle aspiration and outpatient treatment with oral antibiotics are effective (63). The procedure can be performed by an otolaryngologist, an oral surgeon, or an experienced emergency room physician. Needle aspiration is unsuccessful in a small percentage of patients, necessitating incision and drainage. Non–group A streptococcal peritonsillar abscesses recur in 10% of cases, and abscess tonsillectomy may be required for repeated recurrences (63).
Epiglottitis
Acute epiglottitis is a life-threatening, rare complication of URIs. The epiglottis serves as a valve that closes over the proximal portion of the trachea during swallowing to prevent aspiration. When the epiglottis becomes inflamed, the resultant edema causes it to curl posteriorly and inferiorly, thereby reducing the glottic aperture. Inspiration, which draws the epiglottis down, further reduces the effective airway. Since the introduction of the H. influenzae B vaccine, the incidence in children has decreased, and now the incidence in adults is higher than in children (64).
The diagnosis of epiglottitis should be suspected in patients with a sore throat, odynophagia, and muffled voice, all of short duration. Only one half of patients are febrile and show evidence of pharyngitis, and only one third of patients have cervical lymphadenopathy. Sitting erect, complaint of dyspnea, and stridor noted on inspiration are indications of airway obstruction. Soft tissue radiographs of the neck may show edema of the epiglottis and narrowing of the aperture. The diagnosis is confirmed by indirect laryngoscopy, which reveals marked edema of the epiglottis and supraglottic tissue. This procedure must be performed only in circumstances in which emergency intubation can be carried out, because it may induce (rarely in adults) additional respiratory obstruction.
Management
Patients with suspected epiglottitis should be admitted to the hospital for observation; those with signs of airway obstruction should be admitted to an intensive care unit, where close observation and emergency tracheotomy, if necessary, are possible (64). The remainder of the cases can be managed conservatively in a general ward with antibiotic treatment for the most common organisms, including S. aureus, H. influenzae, S. pneumoniae, and S. pyogenes. Use of topical or systemic corticosteroids does not prevent airway obstruction.
Telephone Assessment and Self-Care for Upper Respiratory Tract Infection
Most clinicians welcome the opportunity to assess URI symptoms initially by telephone. The telephone assessment should accomplish the following:
The following symptoms and signs should be sought: symptoms lasting longer than 3 weeks; fever lasting longer than 1 week or associated with delirium; purulent nasal discharge with sinus pain; purulent sputum, chest pain, dyspnea, or hemoptysis; ear pain or discharge; sore throat and a history of rheumatic fever; the combination of cough and fever higher than 102°F (39°C); hoarseness for longer
P.490
than 1 month; pleuritic chest pain; marked odynophagia; and dysphagia, stridor, and difficulty in breathing.
During influenza outbreaks (November through April), it is important to remember that the abrupt onset of fever and cough are fairly sensitive and specific for influenza infection, and that drug treatment for influenza has been shown to be effective only if started within the first 1 to 2 days of illness.
For patients not needing an office visit, simple instructions for self-care can be provided based on the measures described previously.
Lower Respiratory Tract Infections
The cardinal manifestations of lower respiratory tract infections (LRI) are cough and dyspnea, often accompanied by auscultatory evidence of lower respiratory tract inflammation (rhonchi, rales, wheezes, signs of consolidation). Bronchitis and pneumonia are the major infectious syndromes of the lower respiratory tract in adults. A number of noninfectious conditions can also cause a cough that may be confused with a lower respiratory illness. Table 33.8 summarizes the differential diagnosis of acute and persistent cough (see also Chapter 59).
Acute Bronchitis
Acute bronchitis is one of the most common clinical syndromes encountered in outpatient practice. It is defined clinically as an acute illness characterized by a cough, which is usually accompanied by sputum production, and occasionally by fever and pleuritic chest pain. It is differentiated from pneumonia by the absence of dyspnea, a relatively normal chest examination, and lack of abnormalities on chest radiography. Bronchitis appears to be caused by inflammation of the tracheobronchial tree followed by tracheobronchial hypersensitivity that results in a cough of 1 to 3 weeks’ duration. A cough that continues for longer than 3 weeks is generally referred to as a “persistent” or “chronic” cough. “Chronic bronchitis” is defined as an illness characterized by daily productive cough for at least 3 months in two or more consecutive years in the absence of any other illness that may account for these symptoms (see also Chapter 60 for a discussion of COPD).
Epidemiology
Acute bronchitis in otherwise healthy adults is generally caused by viral agents, including influenza A and B, parainfluenza, respiratory syncytial virus, rhinovirus, adenovirus, and coronavirus (65). Less commonly, acute bronchitis may be caused by bacterial agents such asBordetella pertussis, M. pneumoniae, or C. pneumoniae. Other bacterial pathogens, such as S. pneumoniae, H. influenzae, or M. catarrhalisgenerally do not cause bronchitis in persons without underlying lung disease; however, they may be responsible for bacterial superinfections after an acute viral respiratory illness. Bacterial superinfection is rare in otherwise healthy adults and most commonly occurs in elderly persons and those with underlying chronic medical illness, especially chronic heart and lung disease.
|
TABLE 33.8 Conditions and Agents That Can Cause Acute or Persistent Cough |
|
|
Diagnosis
Patients with a cough as the predominant or only respiratory symptom may have pneumonia, bronchitis, or one of a variety of noninfectious conditions associated with persistent cough. Mucoid sputum production develops in many cases and is not helpful in distinguishing etiologic agents. Diagnostic efforts should be directed at identifying patients with pneumonia and those with noninfectious causes of cough, leaving acute bronchitis as a diagnosis of exclusion. For a primary care provider evaluating a patient with an acute illness characterized by cough, the most important question is often whether chest radiography is indicated for further evaluation. No single symptom, sign, or
P.491
constellation of symptoms and signs can predict pneumonia reliably. However, one review suggested that the likelihood of pneumonia is diminished by the absence of signs of focal consolidation on chest examination or of any of the following vital sign abnormalities: heart rate greater than 100 beats/per minute, respiratory rate greater than 24 breaths/per minute, and oral body temperature greater than 100.4°F (38°C) (66). A chest radiograph usually is not necessary if none of these signs are present unless it is indicated for evaluation of dyspnea or an abnormal physical examination of the chest. Gram stain and bacterial cultures of sputum are not useful in the evaluation of acute bronchitis because most cases are of viral etiology and the sputum is readily contaminated by nasopharyngeal flora.
Treatment
The CDC and the American College of Physicians (ACP) recommend against antibiotics for healthy adults with uncomplicated bronchitis (70). A meta-analysis of eight randomized controlled trials of antibiotics (erythromycin, doxycycline, trimethoprim/sulfamethoxazole) for bronchitis found a modest improvement in symptom duration with antibiotics (one-half day, on average) (71), but when side effects, costs, and the increasing problem of antibacterial resistance are taken into account, the risks associated with treatment appear to outweigh the potential benefits. Not surprisingly, antibiotics are more beneficial in patients with suspected acute bronchitis who actually have an underlying pneumonia and less beneficial in patients with only a simple URI (72). The beneficial effects of antibiotics for cigarette smokers appear to be the same or less than for nonsmokers (73). Sputum color is not necessarily an accurate indicator of purulence. If pertussis is suspected, appropriate diagnostic studies should be performed rather than using presumptive therapy with antibiotics, unless the patient is in an epidemic area. Treatment for pertussis is either erythromycin 500 mg orally four times daily or a standard course of one of the newer macrolides (azithromycin or clarithromycin). Trimethoprim (160 mg)/ sulfamethoxazole (800 mg) orally twice daily for 14 days is an alternative for patients who are allergic to erythro-mycin. See the section Flu Syndrome earlier in this chapter for treatment for treatment options for influenza.
Treatment of cough and systemic symptoms such as fever, myalgias, malaise, and chest pain is generally symptomatic. Cough suppression may be achieved with dextromethorphan but often requires codeine sulfate, 15 to 30 mg every 4 to 6 hours, especially at bedtime (seeChapter 59). Bronchodilator treatment with a β-agonists is not helpful in patients with acute bronchitis and no underlying lung disease, (74) but is indicated in patients with evidence of bronchospasm (such as wheezing) on examination (65). Inhaled corticosteroids have not been shown to have a role in the treatment of acute bronchitis in patients without underlying lung disease. Chinese herbs may help shorten the duration of episodes of acute bronchitis; however, available trial data are limited in size and quality (75). Conventional wisdom has suggested that antihistamines should be avoided to prevent inspissated secretions, although empiric data are lacking. Encouragement of good oral hydration is appropriate for all patients with respiratory tract infection.
Smokers with acute bronchitis should be strongly encouraged to stop smoking, particularly for the duration of the acute illness. Smokers with a history of chronic cough before their bronchitis may be more motivated to discontinue smoking permanently in the face of the acute illness. In 50% of those who discontinue smoking, the chronic cough resolves completely within 1 month. Chapter 27 describes approaches to smoking cessation. There is no evidence that smokers who have not developed COPD will benefit from antibiotic treatment for acute bronchitis (73). Chapter 60 discusses the treatment of acute exacerbations of COPD.
P.492
Prevention
An acellular pertussis vaccine combined with an adult formulation of diphtheria, tetanus toxoid, and acellular pertussis (DTaP) has been shown to be effective and safe (68). It has been proposed that adolescents routinely receive this vaccine in the future, and it may replace the adult tetanus-diphtheria (Td) vaccine for periodic booster immunization in adults.
Pneumonia
More than 3 million episodes of pneumonia occur annually in the United States, and it is responsible for more than 30 million days of disability requiring bed rest and for 600,000 hospitalizations. With influenza, pneumonia ranks seventh among all diseases as a cause of death (76) and first among infectious diseases. Its mortality rate has been rising during the last two decades (77).
Definition and Distinction from Bronchitis
Pneumonia is a LRI usually accompanied by fever, dyspnea, cough, and evidence of consolidation on chest radiography. Cough may be purulent or dry, and associated with malaise, pleuritic chest pain, and constitutional symptoms. Pneumonia is caused by bacterial, fungal, or viral infection of the alveoli, which results in airspace edema and consolidation usually discernable on physical examination and chest radiograph. The resultant arterial-alveolar gradient causes dyspnea. Patients who are early in the disease course of pneumonia or who are dehydrated and patients with emphysema and reduced lung parenchyma may fail to show any infiltrate or may show a patchy infiltrate on their chest film despite the presence of considerable inflammation. Antibiotics are indicated in pneumonia whereas they are not indicated in acute bronchitis, which typically involves a viral infection of the bronchial epithelium.
|
TABLE 33.9 Epidemiologic Characteristics Associated with Specific Community-Acquired Pneumonia Pathogens |
||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||
Pneumonia Syndromes and Causes
Important clues for the etiologic diagnosis of pneumonia may be obtained from knowledge of the seasonal, environmental, and occupational predilections of the various agents that cause pneumonias. Table 33.9 lists pathogens associated with different epidemiologic characteristics of patients.
Bacterial pneumonias make up the majority of all adult pneumonias, and the largest fraction of these are caused by S. pneumoniae (78). Pneumococcal pneumonia may occur in a previously healthy adult, or after a URI, usually with the abrupt onset of shaking chills, fever, pleuritic chest pain, and cough productive of purulent or rusty sputum. Patients with compromised pulmonary clearance of secretions (e.g., depressed consciousness, morbid obesity, abdominal surgery, chronic bronchitis, congestive heart failure, alcoholism) are predisposed to pneumococcal and other bacterial pneumonias; the onset of clinical symptoms may be more insidious in these patients.
Drug-resistant Streptococcus pneumoniae (DRSP) is increasingly a problem worldwide, with up to one half of isolates showing in vitroevidence of resistance (79). Pneumococci resistant to penicillin are often resistant to cephalosporins, macrolides, doxycycline, and trimethoprim/sulfamethoxazole as well. Characteristics of patients at higher risk for DRSP include age older than 65 years, β-lactam therapy in the previous 3 months, alcoholism,
P.493
immunosuppressive illness or medication (including corticosteroid therapy), multiple medical comorbidities, and exposure to a child in a day care center (78). The clinical significance of DRSP pneumonia is still not completely understood. There is some evidence of increased morbidity and mortality among patients infected with DRSP with high levels of resistance, but infections with intermediate resistance appear to respond well to β-lactam treatment for pneumonia and may be of clinical significance only in the treatment of otitis media or meningitis (80).
The so-called atypical pneumonia syndrome, most common among patients younger than 40 years of age, is characterized by a prodrome of headache and myalgia preceding the onset of respiratory symptoms. Respiratory pathogens commonly causing “atypical pneumonia” includeM. pneumoniae, Legionella pneumophila, C. pneumoniae (TWAR strain), a number of viruses (influenza A and B, respiratory syncytial virus, parainfluenza, adenovirus), Chlamydia psittaci, Coxiella burnetii (Q fever), Coccidioides immitis, and Pneumocystis carinii. Nonetheless, all of these pathogens may result in a pneumonitis that is clinically and radiographically indistinguishable from pneumococcal pneumonia (81,82); because “typical” bacterial pneumonias are also commonly preceded by a prodrome of headache, myalgias, and malaise, the designation “atypical pneumonia” is of little use in practice.
Pneumonias caused by Mycoplasma or Chlamydia often manifest with sore throat, as well as fever and cough. C. pneumoniae may cause a biphasic illness, with severe pharyngitis and laryngitis in the first phase, followed by pneumonia (83,84).
Complications of Mycoplasma pneumonia include sinusitis, otitis media, myringitis (diagnostic if bullae are seen), erythema multiforme or erythema nodosum, intravascular hemolysis, meningoencephalitis, toxic psychosis, myocarditis, and pericarditis. Fulminant infection can occur irrespective of age or host status. Persistent hacking cough, lasting as long as 6 weeks despite therapy, is common and requires symptomatic relief with codeine (see Chapter 59). Relapse of the primary disease occurs in up to 10% of cases, usually 2 to 3 weeks after the initial illness, and is probably related to the fact that mycoplasma persists in bronchial epithelium for up to 14 weeks.
Pseudomonas aeruginosa is an uncommon cause of pneumonia, particularly in the outpatient setting. Risk factors for development of P. aeruginosa pneumonia include structural lung disease (e.g., bronchiectasis, cystic fibrosis), corticosteroid therapy, malnutrition, and recent broad-spectrum antibiotic therapy for longer than 7 days. A sputum culture that yields P. aeruginosa is not always indicative of true infection and may represent only colonization.
Legionnaire disease (caused by Legionella pneumophila and other species), in comparison with other causes of pneumonia, is more likely to be associated with headache, confusion, and diarrhea and is less likely to cause cough, expectoration, and thoracic pain (85). Laboratory abnormalities associated with Legionnaire disease include hyponatremia and elevated serum creatine kinase activity. However, many of these symptoms, signs, and laboratory abnormalities may be seen with other typical and atypical bacterial pneumonias.
Pneumocystis carinii pneumonia (PCP) should be considered if the onset of fever, cough, and dyspnea is insidious over 1 to 4 weeks and the patient is immunocompromised or has risk factors for HIV infection (see Chapter 39). The chest radiograph typically shows diffuse bilateral infiltrates but may show focal infiltrates, cysts, pneumothorax, or no abnormality (86).
Primary influenza viral pneumonia is an early complication of influenza illness. It occurs predominantly among the elderly and patients with chronic illnesses and occasionally in healthy young adults. Within the first day or two of illness, the dry cough becomes productive and sometimes bloody, and tachypnea and dyspnea may progress rapidly to hypoxia, cyanosis, and delirium. Diffuse rales are present on examination. The chest radiograph often reveals bilateral interstitial infiltrates, but lobar consolidation may also be seen. ARDS may develop. Immediate hospitalization and intensive care are required, but the mortality rate remains high. Some patients have milder influenza pneumonia, with persistent fever, cough, dyspnea, localized rales, and normal white blood cell (WBC) count, and they subsequently experience a benign course. Although there are no reported studies of antiviral drugs (Table 33.3) for the treatment of influenza pneumonia, their use is reasonable in patients in whom it is suspected.
Influenza may also be complicated by secondary bacterial pneumonia and bronchitis, more commonly in the elderly and in patients with chronic pulmonary or cardiac disease. The presentation is typically biphasic; initial respiratory symptoms are followed by several days of clinical improvement, and then there is an exacerbation of fever with production of purulent or bloody sputum. The predominant bacterial pathogens are S. pneumoniae, H. influenzae, GABHS species, and S. aureus; S. aureus pneumonia has a mortality rate of approximately 50% in this setting.
Hantavirus pulmonary syndrome (HPS) is a rare pneumonitis with a high mortality rate (87), now recognized throughout the United States and western Canada. After 1 to 7 days of nonspecific viral prodrome of fever, myalgias, chills, and headache, the respiratory phase is heralded by dry cough and dyspnea, with rapid onset of pulmonary edema caused by a capillary leak syndrome. The most lethal complication is cardiogenic shock.
Severe acute respiratory syndrome (SARS) was transmitted globally by SARS-associated coronavirus between February and July 2003, causing more than 8,000 cases and 800 deaths. SARS was thereafter contained, but it is
P.494
unclear if or when another outbreak may occur. Continued surveillance is therefore necessary. SARS is characterized by an incubation period of 2 to 10 days; early systemic symptoms followed within 2 to 7 days by dry cough and/or shortness of breath, often without upper respiratory tract symptoms; development of radiographically confirmed pneumonia by day 7 to 10 of illness; lymphopenia in most cases; and recent travel to China or surrounding countries or occupational exposure (88). Further information can be found athttp://www.cdc.gov/ncidod/sars/clinicians.htm.
Evaluation
Physical Examination
No specific signs or constellation of signs can confirm the diagnosis of pneumonia (66); however the physical examination is helpful in distinguishing pneumonia from acute bronchitis, from other causes of dyspnea, for evidence of chronic lung disease and for early detection of complications (e.g., pleural effusion). The physical examination cannot reliably distinguish between bacterial and atypical pneumonia syndromes. Crepitant rales that do not clear with cough are suggestive of pneumonia of either type, but signs of consolidation (increased tactile fremitus, dullness to percussion, bronchial breath sounds, and egophony) are more common in typical bacterial pneumonia. In the early stages of pneumonia, the examination may be normal despite an infiltrate on the chest film. Alternatively, rales and rhonchi may indicate pneumonia before the appearance of an infiltrate.
Diagnostic Testing
Every patient with suspected pneumonia should have a chest radiograph to establish the diagnosis and evaluate for possible complications (78). Although chest radiography is essential for the firm diagnosis of pneumonia, a normal film does not necessarily rule out pneumonia, and radiographic patterns are not a specific indication of the cause (89). For patients with underlying heart or lung disease, measurement of oxygenation by pulse oximetry (or arterial blood gas analysis) are recommended to help to determine the need for hospitalization. Routine blood tests, including complete blood counts and serum chemistries, are of little value in determining the cause of the pneumonia but may have prognostic value and can also play a role in determining the need for hospitalization (78).
A sputum Gram stain is generally not recommended for the evaluation of outpatients with pneumonia (78). It can be helpful occasionally in directing initial therapy, but discordant results between Gram stain and culture in pneumococcal pneumonia, and the lack of diagnostic data for many other common respiratory pathogens including Legionella, Mycoplasma, and Chlamydia, render the Gram stain of limited utility.Sputum cultures likewise have limited utility in the management of ambulatory pneumonias, because sputum samples are often contaminated by oral flora and cultures are often negative if any prior antibiotic therapy has been administered (90).
Serologic testing and cold agglutinin measurement are usually not useful in the initial evaluation of patients with CAP; consensus guidelines recommend against their use to direct therapy (78). Few pathogens can be diagnosed by serologic study of the acute serum specimen; exceptions include pertussis and hantavirus infection. Rapid test kits are available for the office diagnosis of influenza, but there are limitations to their use (see Flu Syndrome). During an apparent community outbreak of a respiratory illness caused by unculturable agents, it is helpful to the public health authorities for practitioners to collect and save acute and convalescent sera for later study at reference laboratories. Guided by epidemiologic clues (Table 33.8), measurement of acute and convalescent titers for antibodies against selected infectious agents may be considered. Testing for antibodies to M. pneumoniae, Q fever, psittacosis, influenza, Legionella species, tularemia,C. immitis, and Histoplasma capsulatum is available at most state diagnostic laboratories and many commercial laboratories.
Management
An etiologic diagnosis of pneumonia would require many different tests and a specific etiology cannot be defined in up to 50% of cases despite extensive diagnostic testing (91). Therefore, antibiotic choices are driven by the epidemiology of host–microbial interaction and treatment decisions focus on the need for hospitalization. The desire to avoid unnecessary hospitalizations has led to studies defining low-risk patients with CAP who can be treated as outpatients (92).
Decision about Hospitalization
The 2003 update to the pneumonia guidelines of the Infectious Disease Society of America suggest a three-step assessment of the need for hospitalization in a patient with pneumonia (93). First, the social setting of the patient should be assessed to determine if it is stable enough to allow reliable oral antibiotic therapy and convalescence. If sufficiently stable, the second step is calculation of the Pneumonia Severity Index (PSI). Third, the PSI should be supplemented as appropriate with clinical judgment.
The PSI originated from a meta-analysis of pneumonia outcomes, which revealed multiple prognostic factors associated with increased mortality: altered mental status, male gender, absence of pleuritic chest pain, hypothermia, systolic hypotension, tachypnea, diabetes mellitus, neoplastic disease, leukopenia, and multilobar pulmonary infiltrates (94). A survey of practitioners making decisions on hospitalization for pneumonia indicated that hypoxemia, inability to maintain oral intake, and lack of patient home
P.495
care support were almost universal and appropriate criteria for admission. However, practitioners usually overestimated the risk of death from pneumonia, based on examination and comorbidity, resulting in excessive use of hospitalization (95). The large Patient Outcomes Research Team (PORT) study, which used data from more than 14,000 patients with CAP, established that if the answers to all of the following three questions are “No,” the patient is in the lowest of five risk classes (group I), with a 30-day mortality rate less than 0.4% (92):
If the answers to any of these three questions was “Yes,” further risk assessment was determined by a point scoring system including the additional characteristics listed in Table 33.10, each of which was independently associated with increased morbidity. Patients in risk class I had a 0.1% to 0.4% risk of mortality within 30 days; class II patients (70 points or less) had a 0.6% to 0.7% risk, and class III patients (71 to 90 points) had a 0.9% to 2.8% risk. (92). Most patients in classes I and II may be safely managed as outpatients; class III patients are potential candidates for outpatient therapy or a brief inpatient observation. In contrast, patients in classes IV (91 to 130 points) and V (more than 130 points) had 30-day mortality rates of 8.2% to 9.3% and 27.0% to 31.1%, respectively, and are usually best treated in hospital settings. The PORT-derived indicators of risk can be used as general guidelines until prospective studies provide even more precise rules, but clinical judgment should supersede such rules. Complications such as concomitant meningitis or septic arthritis, hemoptysis, prior splenectomy, and need for respiratory isolation of potential tuberculosis or pneumonic plague must be considered. Moreover, a patient's social situation and personal preferences should play a role in the decision whether to hospitalize.
Ambulatory Management: Choice and Duration of Antimicrobial Therapy
The treatment of CAP is by necessity empiric and based on a knowledge of epidemiology and estimates of patient risk for different pathogens.
The American Thoracic Society has issued consensus guidelines for the initial treatment of immunocompetent adults with CAP (78); these guidelines are reasonable and practical but require further study to show that adherence improves outcome. The guidelines divide ambulatory patients into two groups, based on the presence or absence of cardiopulmonary disease and modifying factors that increase the risk of infection with drug-resistant pneumococcus, enteric gram-negative bacteria, or P. aeruginosa (Table 33.11).
|
TABLE 33.10 Point Scoring System for Step 2 of the Prediction Rule for Assignment to Risk Classes II (=70 Points), III (71–90 Points), IV, and V |
||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||
For ambulatory patients with no comorbidity and no risk factors for drug-resistant S. pneumoniae or other modifying factors, an advanced-generation macrolide (azithromycin or clarithromycin) is recommended as first-line therapy; doxycycline is an alternative for patients who are allergic to or intolerant of the macrolides (Table 33.12). Clarithromycin and azithromycin provide coverage for pneumococcal, chlamydial, Legionella, and mycoplasmal infections. Azithromycin has the advantage of once-daily and single dosing. Clarithromycin, but not azithromycin, can raise blood theophylline levels, occasionally into the toxic range, and dosages of the latter drug should be monitored and adjusted.
|
TABLE 33.11 Modifying Factors That Increase the Risk of Infection with Specific Pneumonia Pathogens |
||
|
P.496
For ambulatory patients with either cardiopulmonary illness (congestive heart failure or COPD), a risk factor for drug resistant S. pneumoniae, or another modifying factor, the guidelines recommend treatment with a β-lactam plus a macrolide or doxycycline (Table 33.13). The β-lactam options include oral cefpodoxime, cefuroxime, high-dose amoxicillin (1 g every 8 hours), amoxicillin/clavulanate, or parenteral ceftriaxone followed by cefpodoxime. An alternative is an antipneumococcal fluoroquinolone (gatifloxacin, levofloxacin, moxifloxacin) that is active against S. pneumoniae, M. pneumoniae, Chlamydia trachomatis, Legionella, M. catarrhalis, and gram-negative aerobes. See Table 33.3 for dosing guidelines for these drugs. The fluoroquinolones are generally effective against DRSP, although resistance has been reported (96) and is increasing due to over-use of these agents. Many experts caution against the indiscriminate use of fluoroquinolones for outpatient treatment of pneumonia, because of concerns about the emerging resistance of pneumococci to this class of antibiotics (80).
|
TABLE 33.12 Outpatient Treatment of Pneumonia, Group I: Patients with No Cardiopulmonary Disease or Modifying Factorsa |
||||||||
|
||||||||
|
TABLE 33.13 Outpatient Treatment of Pneumonia, Group II: Patients with Cardiopulmonary Disease or Other Modifying Factorsa |
||||||||
|
||||||||
The duration of therapy for CAP is not precisely defined. The American Thoracic Society guidelines suggest that “typical” bacterial infections, such as pneumococcal pneumonia, should be treated for 7 to 10 days, whereas Mycoplasma, Chlamydia, and Legionella pneumonias may need longer therapy, ranging from 10 to 14 days. Because an etiologic diagnosis usually is not established in outpatients, clinicians should consider the presence of coexisting illness, the severity of illness at the onset of antibiotic therapy, and the subsequent course in determining the duration of antibiotic therapy. For most infections, a 5-day
P.497
course of azithromycin is adequate therapy because of its prolonged biologic half-life.
Followup
The patient should be advised to keep in close contact by telephone, maintain good hydration with oral fluids, use aspirin or acetaminophen to control fever and headache, and avoid cigarettes. A telephone contact with the patient 24 hours after the initial visit provides a check on antibiotic adherence and side effects and on the status of symptoms; also it reassures the patient that he or she has access to the clinician should the condition worsen or fail to improve.
A followup visit to the office 3 to 4 days later will help assess response to therapy. Symptoms of pneumococcal pneumonia in the uncompromised host usually abate within 48 to 72 hours after initiation of therapy, and somewhat longer with other pathogens or compromised host defenses. If a substantial clinical response to the initial antibiotic therapy has not occurred in this time, the patient must be reevaluated. Possible reasons for clinical failure include poor adherence to the antibiotic regimen; resistance of the etiologic organism to the empiric antibiotics prescribed; unusual pathogens such as tuberculosis, viral, or fungal pneumonia; and a noninfectious cause such as pulmonary embolus or carcinoma. In any case, hospitalization usually is required to determine the cause of therapeutic failure and to provide additional treatment. Complete resolution of symptoms caused by an episode of pneumonia may not occur for 30 days or longer after diagnosis (96).
After clinical resolution of the pneumonia, a chest radiograph is recommended in 4 to 6 weeks to exclude malignancy or other persistent lung abnormalities, particularly in smokers and in patients older than 40 years of age (78). The rate of radiographic resolution depends on age and extent of pneumonic involvement; although most patients have complete clearance at 4 weeks, elderly patients and those with multilobar pneumonia or underlying lung disease can have delayed resolution (97).
Prevention of Pneumonia
Chapter 18 discusses polyvalent pneumococcal and influenza vaccines in detail. No special precautions need be taken to isolate the ambulatory patient with pneumonia. Household contacts of these patients need no special surveillance, with the exceptions of pneumonic disease caused by tuberculosis (see Chapter 34), tularemia, plague, or meningococci.
Pleuritis and Pleurodynia
Pneumonic infections may cause inflammation or infection of the pleura, resulting in pleuritic chest pain and the appearance of a pleural effusion. Parapneumonic effusions and empyemas are typically caused by bacterial pneumonias, but tuberculosis, atypical bacteria, viruses, fungi, and even parasites may cause pleuritis. Chapter 59 discusses the evaluation of pleural effusions.
Pleurodynia is an uncommon acute illness usually caused by one of the coxsackie viruses. It occurs in summer and early fall. The presenting symptoms may suggest the onset of pneumonia: abrupt onset of severe paroxysmal pain of the thorax or abdomen, worse with cough or breathing. Other manifestations of pleurodynia include fever, headache, cough, and anorexia. The physical examination is often normal except that the patient splints to avoid pain, which is commonly felt in the lower rib cage or under the sternum. The chest radiograph is usually normal. Most patients recover within 3 days to 1 week. Rare complications are orchitis, pericarditis, and aseptic meningitis.
Appropriate Prescribing of Antibiotics for Respiratory Tract Infections
Antibiotics for respiratory tract infections account for approximately 50% of antibiotic prescriptions written in physicians’ offices in the United States (98). Although there was a decline in the number of antibiotic prescriptions for respiratory tract infections during the last decade, approximately 50% of patients with colds or nonspecific URIs, 60% of patients with acute bronchitis, and 70% of patients with sinusitis still received antibiotics (98, 99, 100). More than half of patients who were prescribed an antibiotic for these URIs were treated with a broad-spectrum agent (100). A comparison of these rates with the estimated incidence in adults of acute bacterial sinusitis (15%) and GABHS pharyngitis (10%)—the principal acute URIs for which antibiotics are indicated—strongly suggests that the use of antibiotics for URIs in practice should be reduced.
The overuse of antibiotics for respiratory tract infections exposes patients to unnecessary side effects, especially allergic reactions, diarrhea, and vaginitis (Table 33.3); increases medical care costs both for individual patients and health care payors; and contributes to the increasingly serious problem of antibiotic resistance.
Antibiotic Resistance
The importance of emerging antibiotic resistance in relation to respiratory tract infections is illustrated by the increasing resistance of S. pneumoniae to multiple antibiotics. Since the 1990s, the rates of resistant strains of this key respiratory pathogen have increased from 25% to 40% for penicillin, 29% to 35% for trimethoprim/sulfamethoxazole, 8% to 20% for tetracyclines and 20% to 28% for macrolide antibiotics (101, 102, 103, 104). Although the most common respiratory bacterial pathogens remain susceptible to the newer fluoroquinolones (Table 33.3),
P.498
increasing resistance has been reported for older ones such as ofloxacin, with rates of about 5% (105).
The proliferation of day care arrangements for young children and the increase in global travel also contribute to the emergence of antibiotic resistance, but the frequent prescription of antibiotics for respiratory tract infections probably is a major contributor to this problem. There is evidence that individual patients who received multiple or recent courses of antibiotics are more likely to be colonized with resistant strains of bacteria. It is also clear that rising rates of antibiotic resistance for a particular antibiotic parallel increased use of that antibiotic; efforts to decrease the use of certain antibiotics in some countries have resulted in declining rates of resistance to those antibiotics (106). Although the possibility of resistant bacteria in some URIs, such as suspected acute bacterial sinusitis, warrants the consideration of broad-spectrum antibiotics in selected patients, their indiscriminate use will only worsen the problem of resistance.
Successful Approaches to Limiting Antibiotics
Practitioners can use the following principles to limit antibiotic prescriptions for respiratory tract infections to appropriate indications:
Similar principles have been applied successfully in practice settings (70,107). A multifaceted educational intervention targeted at patients and clinicians successfully reduced antibiotic prescriptions for bronchitis from 74% to 48%, without increasing the duration of illnesses or utilization of medical services or other medications (108). An important component of that intervention was use of the term “chest cold” instead of “bronchitis” in referring to cough illness. This study and others emphasize the importance of providing adequate explanations to patients of the appropriate use of antibiotics.
A successful approach to patients seeking care for respiratory tract infections for whom antibiotics are not indicated should include the following:
Contrary to the beliefs of many clinicians, patients’ expectations for antibiotics may not be insurmountable. Reduced prescribing of antibiotics is not necessarily associated with less-satisfied patients (70). Rather, patient satisfaction with encounters for respiratory complaints may have more to do with the adequacy of explanations for their illness (109). Many governmental, national, and local health care organizations are working to reduce antibiotic use for respiratory tract infections, and clinicians should increasingly be able to find support in this effort. The CDC's Campaign for Appropriate Antibiotic Use in the Community provides some helpful resources and materials for clinicians (http://www.cdc.gov/drugresistance/community).
Specific References*
For annotated General References and resources related to this chapter, visit http://www.hopkinsbayview.org/PAMreferences.
P.499
P.500