#94674: Pneumonia

Bacterial causes of CAP predominate, accounting for at least half of all adult cases, including older individuals [9,42]. S. pneumoniae is the leading cause of CAP in any adult age-group, with or without comorbid conditions [6,7,10]. It is estimated that pneumococcal infection accounts for 20% to 60% of all hospitalized patients with pneumonia [6]. Common bacterial pathogens other than S. pneumoniae include H. influenzae type b, S. aureus, and gram-negative bacilli [25,26,29,48]. H. influenzae type b is a small, pleomorphic gram-negative rod known for causing pneumonia in older adults and patients with underlying lung disease.

Atypical pneumonia (and the pathogens associated with this syndrome) is so labeled because the onset of illness tends to be subacute and the clinical exam and radiographic features lack the classical findings seen with typical cases of pneumonia. The most common atypical pathogens are Mycoplasma pneumoniae and C. pneumoniae, followed by Legionella spp. [9]. M. pneumoniae is a tiny bacterium that lacks a rigid cell wall. It is spread by droplet nuclei, and transmission within a community proceeds slowly over many weeks. Mycoplasma infection is a disease of adolescence and young adulthood, and it is the most common cause of atypical pneumonia in those younger than 40 years of age [66]. Small cluster outbreaks of pneumonia have been observed in large families, schools, nursing homes, and other closed population. There are about 60 different species of Legionella, but most disease is caused by Legionella pneumophila, a gram-negative rod usually transmitted via inhalation of aerosolized water contaminated with the bacteria [229].

The distribution of etiologic agents varies in relation to illness severity and management setting. In cases of relatively mild illness that permit outpatient treatment, blood cultures are rarely positive and the diagnosis relies on sputum culture and/or serial serology. In a Canadian study of CAP in the ambulatory setting, designed to determine the frequency of usual and atypical bacterial pathogens, an etiologic diagnosis was established in 48% of patients examined [222]. Of the 419 patients who had blood cultures, 7 (1.4%) were positive, all for S. pneumoniae. The atypical pathogen group (M. pneumoniae or C. pneumoniae) accounted for 29% of cases, S. pneumoniae for 6%, and Haemophilus spp. for 5%. The etiologic role of viruses was not studied [222].

A similar distribution and frequency was observed in a well-studied series from Spain, comparing pneumonia microbial etiology in three clinical management settings: outpatient, inpatient on the general care ward, and inpatient admissions to the ICU [29]. Among outpatients with CAP, the most frequently identified etiology was the atypical pathogen group (36%), followed by S. pneumoniae (35%), viruses (9%), and mixed etiologies (9%). As the severity of illness increased, marked by admission to the hospital general ward and ICU, the likelihood of mycoplasma or chlamydia etiology decreased substantially (14%) and the frequency of S. pneumoniae (43%), mixed bacterial pathogens (22%), S. aureus, Pseudomonas, and other gram-negative bacteria infection increased.

S. aureus is an uncommon cause of CAP but should be suspected during influenza outbreaks and in any patient with sepsis syndrome and multifocal pulmonary infiltrates. The role of S. aureus, and MRSA specifically, was examined in an observational study of 627 CAP cases admitted to 12 university-affiliated hospitals during the winter months (influenza season) of 2006–2007 [49]. Of the 595 patients from whom blood and sputum cultures were collected, a bacterial pathogen was identified in 107 (17%). The most common pathogen identified was S. pneumoniae (57 cases), followed by S. aureus (23 cases, 14 of which were MRSA). Thus, S. aureus accounted for 5% of the total and 22% of the cases in which the etiology was identified. Of the 23 patients with staphylococcal pneumonia, blood cultures were positive in 39% and sputum culture in 89%. Clinical features observed to be highly associated with S. aureus infection were multiple pulmonary infiltrates, altered mental status, illness severity requiring ICU admission, and intubation [49].

Studies have indicated that 5% to 20% of adult CAP may be caused by a viral pathogen [50]. However, as noted earlier, the role of respiratory tract viral infection in pneumonia is complex and perhaps underestimated. Studies utilizing newer diagnostic methods such as PCR have demonstrated rates of viral infection as high as 39% in patients presenting with pneumonia [9,34]. Because these studies rely on specimens and washings taken from the nasopharynx, rather than directly from the lung, it is not clear to what extent viral isolates in this setting represent primary pneumonia pathogens or concomitant viral upper respiratory infection that may impair pulmonary defense mechanisms and thus predispose to bacterial pneumonia.

Clinical and pathologic studies of pneumonia during influenza seasons have demonstrated clearly that influenza virus (types A and B) is an important cause of primary viral CAP [25,47]. Seasonal respiratory syncytial virus (RSV), rhinovirus, adenovirus, and parainfluenza virus are also commonly associated with pneumonia in adults [31,34,47]. Since the advent of coronavirus disease 2019 (COVID-19) pandemic, SARS-CoV-2 has superseded the usual viral respiratory pathogens. RSV and rhinovirus are especially common among older adults and nursing home residents [31]. Clinical studies that utilize viral culture for case definition have found that RSV can be recovered from 3% to 10% of older adults with pneumonia [30]. The paramyxovirus hMPV, first isolated in 2001 from children hospitalized with acute respiratory infection, has now been reported in all age groups and stages of disease, from asymptomatic carrier state to severe bronchitis and pneumonia [30].

Mixed viral-bacterial infection has been documented in 30% of adult cases of CAP in some studies [9,31,34]. S. pneumoniae in combination with rhinovirus, influenza A, or RSV is found most commonly [34]. On rare occasions fungal and parasitic pathogens are isolated in association with CAP syndrome.

The clinical recognition of CAP in adults is challenging because its presentation is similar to other acute respiratory illnesses such as pulmonary embolism/infarction and congestive heart failure [3,51,52]. Diagnosis relies on clinical features combined with radiographic findings; however, both the clinical presentation and chest x-ray abnormalities are variable and in part nonspecific, particularly in the elderly [3,29]. Common presenting symptoms and signs are:

Pneumonia in the elderly may present without a history of chills or fever, little cough, and a paucity of findings on exam and chest x-ray. Often in such cases, some combination of tachypnea, tachycardia, and altered mental status is the only sign [31,42].

Physical examination should focus on the chest, with auscultation to detect localized crackles (rales), bronchial breath sounds, and other signs of consolidation or pleural effusion [47]. Pulse oximetry should also be done. The most clinically significant individual findings are (in descending order) egophony, bronchial breath sounds, and dullness on percussion [53].

When pneumonia is suspected on the basis of these clinical features, chest radiography is the standard for confirming the diagnosis, and posteroanterior and lateral radiographs are recommended [3,29]. Some degree of infiltrate is almost always demonstrated on chest radiographs of patients who have been ill longer than 24 to 48 hours, although the appearance may be subtle or absent on initial presentation [29,47]. Pneumonia is described according to its anatomic distribution on chest radiographs as either lobar, multifocal/lobar, bronchopneumonic, or interstitial. Chest radiography also provides clues to alternative diagnoses having similar signs and symptoms. Computed tomography (CT), which is more sensitive than chest radiographs, may be useful in selected patients where diagnostic considerations are complex and initial radiographic studies are negative or inconclusive.

The characteristic symptoms and signs, combined with radiographic findings of an infiltrate, establish the clinical diagnosis of pneumonia. One validated prediction tool commonly used assigns 1 point for each of five clinical features present in conjunction with an infiltrate on chest radiography [54]:

A score of 4 or 5 indicates a 25% to 50% probability of pneumonia; a score of 2 or 3 indicates a probability of 3% to 10%; and a score of 0 or 1 represents a probability of l% or less [29,54]. Neither clinical nor radiographic features can reliably differentiate primary viral from bacterial or combined viral-bacterial pneumonia [9,31,32]. There are some features that, if present, aid in making the distinction. The presence of a viral epidemic in the community, such as influenza or RSV, increases the likelihood of a viral etiology [32]. The patient's age can also help identify the most probable cause; as noted previously, viral infections have been found more often in young children and adults older than 60 years of age compared with younger adults [9,24]. Chest pain is significantly more frequent in adults with bacterial pneumonia than in those with viral pneumonia [9]. Radiographic findings are generally not useful in identifying a specific pathogen, although multilobar infiltrates suggest infection with S. pneumoniae, S. aureus, or L. pneumophila, and patchy, interstitial infiltrates suggest a viral or mycoplasmal etiology [47,49].

The first use of the term atypical pneumonia was in 1938 to describe a series of seven patients who had developed an unusual form of tracheobronchitis [65]. There had also been descriptions of outbreaks of pneumonia that behaved atypically in Europe in the 1920s. In general, these outbreaks were milder and had higher recovery rates than expected for the typical case of pneumonia.

At the present time, atypical pneumonia is encountered, and managed, primarily in the outpatient setting. The causative pathogen most commonly identified in such cases is M. pneumoniae. According to CDC estimates, Mycoplasma infections occur at the rate of 2 million cases each year and are responsible for between 1 and 10 of every 50 cases of CAP [66].

Atypical pneumonia syndrome, best represented by mycoplasma infection, presents with a subacute prodrome of malaise, low-grade fever, headache, myalgia, and non-productive cough. Symptoms progress slowly over days to weeks; often patients are thought to have an upper respiratory infection or bronchitis and appear less ill than those with typical bacterial pneumonia [65,66]. The physical examination usually reveals fine rales but no signs of lung consolidation. In the early stage, there may be maculopapular skin eruptions and, on examination of the ear canal, bullous myringitis of the tympanic membrane. Chest x-ray reveals patchy alveolar densities or inhomogeneous segmental infiltrates, often bilateral involving the middle lobe and lingual. The white blood cell count may be normal or only slightly elevated. Full recovery is expected with no residual effects in a previously healthy individual. However, the disease can be severe in those with sickle cell anemia, older adults, and those with immunosuppression [65].

In younger patients, C. pneumoniae (TWAR strain) infection may present as atypical pneumonia. Outbreaks tend to occur in communal settings such as military units and college dormitories [231]. The illness is similar to that seen with mycoplasma infection, except that laryngitis is a prominent feature and nonexudative pharyngitis is common [26]. Chest x-ray may show patchy consolidation, interstitial infiltrates, or funnel-shaped lesions. The white blood cell count is usually normal.

The first recorded outbreak of legionellosis occurred in 1976 at an annual convention of the American Legion in Philadelphia. A total of 182 of the delegates (many of whom were elderly) became ill, and 146 were hospitalized. The mortality rate was 16%. Because the conference ended prior to the development of significant symptoms in many patients, hospitals all over the United States admitted one or more of the patients who had attended the convention. Despite an outpouring of resources, it took six months to isolate the organism, later named L. pneumophila. The pneumonia caused by the organism is commonly known as Legionnaires' disease [65].

L. pneumophila is a small gram-negative bacillus, atypical in its clinical presentation and for its lack of susceptibility to ß-lactam antibiotics. There are about 60 identified species of Legionella, although L. pneumophila is the primary pulmonary pathogen [230]. Legionella accounts for an estimated 8,000 to 18,000 cases of pneumonia requiring hospitalization in the United States each year [229,230]. Suspicion for infection with Legionella organisms should be high in older adults, in those with chronic underlying disease, and in all patients with pneumonia severe enough to require hospitalization.

Legionella bacteria are found in common sources of freshwater but not usually in sufficient numbers to cause disease. However, in commercial water systems such as those found in large buildings, storage tanks, cooling towers, decorative fountains, or hot tubs, Legionella growth exceeds the threshold required for transmission to susceptible hosts via aerosolization [229]. Because hotels, resorts, and cruise ships often use large, complex water systems and other aerosol-generating devices, travel is a risk factor for disease. This is also true for hospitals and long-term care facilities.

The onset of infection is marked by dry cough, fever of 38.3°C–38.8°C (101°F–102°F), then progressive symptoms and signs of pneumonia accompanied by multi-organ involvement—vomiting, diarrhea, headache, and altered mental status. Chest x-ray reveals rapidly progressive, asymmetric infiltrates without signs of consolidation. Prompt diagnosis relies on clinical suspicion, urine antigen assay, and specialized culture techniques.

The challenge of diagnosis is complicated by the lack of cost-effective, reliable, and rapidly available tests to discriminate between viral and bacterial pneumonia [37]. The IDSA/ATS guideline notes that routine cultures of sputum and blood are not recommended for patients treated in the ambulatory setting, as results rarely impact management decisions [47]. The primary reason for cultures and serologic testing is to identify specific pathogens suspected on the basis of clinical and epidemiologic findings or cases in which the results of testing will substantially alter the empirical treatment of the patient [47]. Testing is recommended when there are symptoms or signs of severe CAP (e.g., multilobar infiltrates, respiratory insufficiency, sepsis, leukopenia), when drug-resistant or unusual pathogens are suspected (e.g., MRSA, Pseudomonas, Legionella), and when patients do not show clinical improvement within 72 hours after starting empiric treatment [235].

Blood Culture

Blood cultures are optional and not recommended as a routine diagnostic test for CAP managed in the ambulatory setting. The principle reason is that the yield is low, and studies show that a positive culture leading to a change in antimicrobial therapy occurs in about 3% or fewer cases [55,56,222]. The IDSA/ATS guideline recommends pretreatment blood cultures in patients managed in the hospital who are classified as severe CAP, or being empirically treated for MRSA or gram-negative bacilli, or have a history of hospitalization and antimicrobial therapy within the previous 90 days [235]. Blood and sputum cultures should also be obtained in patients hospitalized with CAP and any one of the following conditions:

Blood cultures are indicated for patients who have severe CAP, as they are more likely to have infection with a pathogen other than S. pneumoniae [235].

The ATS and the American College of Emergency Physicians (ACEP) also note that blood cultures need not be obtained routinely in all patients admitted with CAP [57]. Similarly to IDSA/ATS, ACEP adds that blood cultures should be considered for patients at higher risk, such as persons who have compromised immune systems, significant comorbidities, severe disease, or another risk factor for infection with resistant organisms [57].

Sputum Culture and Gram Stain

Sputum Gram stain and culture are also considered optional and are not recommended for routine management of adult CAP in the ambulatory setting. Pretreatment Gram stain and culture of respiratory secretions are recommended in adults hospitalized with CAP, in accordance with the same decision criteria outlined above for obtaining blood cultures [235]. Examination and culture of respiratory secretions should be performed only on specimens that meet quality performance measures for collection, transport, and processing of sputum samples.

The diagnostic utility of sputum Gram stain and culture has been demonstrated in patients hospitalized with pneumococcal pneumonia confirmed by positive blood culture. In a series of 58 patients from whom good quality sputum specimens (>10 inflammatory cells per epithelial cell) were submitted before or within six hours after initiation of antibiotic therapy, pneumococci were identified by Gram stain in 63% and by culture in 89% of cases [224].

Newer Diagnostic Techniques

Assays for the detection of antigen and other components of bacterial and viral pathogens have become a useful adjunct for establishing the etiology of pneumonia. Among these is the detection of bacterial antigen in the urine of patients with CAP. In a clinical series report, an assay for S. pneumoniae cell wall polysaccharide in urine was positive in 64% of patients with pneumococcal pneumonia; the sensitivity increased to 88% in patients who were bacteremic [225].

In a meta-analysis of published studies, the assay for detection of Legionella antigen in the urine of patients with pneumonia has been shown to have excellent specificity (99%) but only modest sensitivity (74%) [226]. Thus, a urine Legionella antigen assay is very useful to "rule in" the diagnosis but does not rule it out—a negative result should be interpreted with caution. The ATS/IDSA guideline recommends against routinely testing urine for Legionella antigen in adults with CAP, except when indicated by epidemiologic factors, such as association with a Legionella outbreak or recent travel or in adults with severe CAP [235]. Isolation of Legionella from sputum can be accomplished on selective media. Serologic diagnosis requires acute and convalescent serum; it is useful to confirm a case, but of little value in early diagnosis.

Testing for Viruses

Viral culture remains the criterion standard for diagnosis of viral pneumonia, but because of limitations such as the need for prompt transportation, time needed for viral detection, and the lack of sensitivity for all viruses, rapid antigen testing is often done. In adults, rapid testing has a sensitivity of 50% to 60% and a specificity of at least 90% [31]. Testing of nasal swab specimens is slightly less sensitive than testing of wash specimens, but wash specimens can be difficult to obtain in frail or cognitively impaired adults. Rapid RSV tests are usually not useful for adults, as the level of virus titers shed is low [31]. Diagnostic testing (PCR) for SARS-CoV-2 by nasopharyngeal swab may be performed on patients presenting with CAP in areas experiencing COVID-19 epidemic activity, although lower respiratory symptoms typically begin as upper respiratory symptoms are resolving or have already resolved.

Molecular diagnostic testing of sputum holds promise for providing a rapid and accurate etiologic diagnosis. Studies show that real-time PCR is significantly more sensitive and specific for the detection of the common respiratory viruses that cause CAP, as well as M. pneumoniae and C. pneumoniae [24,33]. However, molecular assays are expensive and not currently widely available [31].

Biomarkers

Over the past several years, researchers have been evaluating biomarkers for their utility in diagnosis and for determining duration of empirical therapy for presumed bacterial pneumonia. Procalcitonin has been shown to be superior to other commonly used markers for its specificity for bacterial infection and its ability to distinguish CAP from asthma and COPD [58,59]. This marker has predictive value; however, no biomarker should be used on its own and, if used, should be considered within the context of clinical and laboratory findings [59]. The 2019 IDSA/ATS guidelines do not recommend the use of procalcitonin to determine need for initial antimicrobial therapy [235].

One of the most important decisions in the management of CAP is determining the site of care—that is, outpatient or inpatient and, if the latter, a general care floor or an ICU [68]. Many physicians admit patients to the hospital when they could be managed effectively on an outpatient basis [47]. This decision requires a careful evaluation of the severity of illness in the context of the personal and social well-being of the patient. Objective severity-of-illness scores and prognostic models can aid in identifying patients who may require hospitalization or admission to an ICU. The most widely used scales are the CRB-65 (confusion, respiratory rate, blood pressure, age 65 years or older) (Figure 1), the CURB-65 severity score (which adds urea level to the CRB-65 criteria), and the Pneumonia Severity Index (PSI) (Table 5). These assessment tools are recommended by the IDSA/ATS as an aid to clinical judgment in determining the site of care [47,69,70]. The scales have been compared, and they do not differ significantly in overall performance [71]. However, each scale has advantages and disadvantages, and none factor in all clinical considerations (such as comorbidities or social factors) [68]. CURB-65 and CRB-65 are easier to score as they have fewer variables and are more likely to correctly classify high-risk patients (i.e., high positive-predictive value) [72]. In contrast, the PSI is more sensitive and is better at determining which patients do not require hospitalization (i.e., low false-negative rate). About 30% to 60% of patients at low risk are unnecessarily admitted to the hospital according to the PSI score [68].

CLINICAL SEVERITY ASSESSMENT IN THE COMMUNITY SETTING: THE CRB-65 SCORE

aDefined as a Mental Test Score of 8 or less or new disorientation in person, place, or time.

Figure 1

Source: Reprinted with Permission from Lim W, van der Eerden MM, Laing R, et al. Defining community-acquired pneumonia severity on presentation to hospital: an international derivation and validation study. Thorax. 2003;58:377-382.

The PSI, CURB-65, and CRB-65 were developed to predict the risk of death. Because this risk does not always equate to the need for hospitalization and/or ICU admission, other scales have been developed. For example, SMART-COP provides a score based on a composite of systolic blood pressure, multilobar involvement on chest radiograph, albumin level, respiratory rate, tachycardia, confusion, oxygenation, and arterial pH [73]. SMART-COP was found to accurately predict the need for intensive respiratory or vasopressor support. Another tool, the Severe Community-Acquired Pneumonia (SCAP) score, includes points assigned to eight variables: arterial pH, systolic pressure, confusion, blood urea nitrogen level, respiratory rate, chest radiograph findings, pulmonary arterial oxygen tension (PaO₂), and age (older than 80 years) [74]. SCAP has identified a larger proportion of patients as low risk compared with the PSI, CURB-65, and CRB-65, and is better than or as accurate as those scores at predicting adverse outcomes in hospitalized patients [74,75]. The IDSA/ATS guideline notes that the results of these objective criteria should always be accompanied by clinical judgment, including consideration of subjective factors, such as the availability of outpatient support resources and the patient's ability to safely and reliably take oral medication [47].

It is estimated that admission to an ICU is needed for 10% to 20% of patients hospitalized with CAP [76]. The IDSA/ATS guideline recommends two major and nine minor criteria to define severe pneumonia requiring ICU admission. [47]. The major criteria are septic shock requiring vasopressors or acute respiratory failure requiring intubation and mechanical ventilation. The presence of at least three of the following minor criteria suggests the need for ICU admission [47]:

These criteria are based on empirical evidence from published studies and validated in pneumonia cohorts from different countries [235,77,78].

The goal of antibiotic treatment of pneumonia is to eradicate the infection or to reduce the bacterial load so the patient's own immune response is able to limit spread and speed recovery. The choice of regimen is based on consideration of known or suspected etiology, patient's age and severity of illness, comorbidities, and knowledge of resistance patterns in the community. When culture results are known, one should tailor therapy in accordance with antibiotic sensitivities and avoid unnecessarily prolonged treatment so as to minimize the potential for the development of resistance [37].

Pending results of cultures and serologic testing, an initial empiric treatment regimen is selected according to patient variables and clinical setting (Table 6) [47]. Patients with mild illness and no serious coexisting disease may be managed as outpatients. The 2019 ATS/IDSA guideline recommends amoxicillin 1 g three times daily, doxycycline 100 mg twice daily, or a macrolide (e.g., azithromycin 500 mg on first day then 250 mg daily or clarithromycin 500 mg twice daily) [235]. The macrolide monotherapy recommendation is conditional based on prevalence of local pneumococcal resistance (<25%) and provided the patient has not received antimicrobials within the previous three months [47]. S. pneumoniae resistance to macrolides is four times more likely in adult patients who have received this class of drug within the previous three months, in which case a fluoroquinolone or ß-lactam plus macrolide combination should be selected. Patients with comorbidities should receive broader spectrum treatment as they are more likely to harbor resistant pathogens and to be more vulnerable to poor outcomes if the initial regimen is inadequate. For outpatient adults with comorbidities, the ATS/IDSA guideline recommends one of the following options (in no order of preference) [235]:

The American College of Radiology asserts chest x-ray is the imaging modality of choice for complicated pneumonia.

(https://acsearch.acr.org/docs/69446/Narrative Last Accessed: August 29, 2024)

Strength of Recommendation: 9

The choice of treatment option requires consideration of clinical and epidemiologic factors unique to patient and locale. As a rule, patients with recent exposure to one class of antibiotics should receive treatment with antibiotics from a different class to avoid possible bacterial resistance to the initial regimen. Fluoroquinolones should not be used routinely, as widespread use increases the possibility that resistance will develop.

The selection of a respiratory fluoroquinolone or a ß-lactam plus macrolide combination is recommended also for patients with CAP who are hospitalized on a general floor [235]. Patients with recent hospitalization and parenteral antibiotic treatment, as well as those with severe pneumonia usually requiring admission to an ICU, need empiric treatment for usual pathogens and consideration of coverage for S. aureus and gram-negative bacteria infection pending sputum and blood culture results. This is achieved with a regimen that combines a broad-spectrum ß-lactam (e.g., piperacillin/tazobactam) or a carbapenem with either azithromycin or a respiratory fluoroquinolone, adding vancomycin or linezolid to cover MRSA if there is clinical suspicion of S. aureus infection. Antibiotic selection for treatment of CAP in reference to specific pathogens is summarized in Table 7 [47].

The Infectious Diseases Society of America recommends empirical therapy for MRSA pending sputum and/or blood culture results for hospitalized patients with severe community-acquired pneumonia defined by any one of the following: a requirement for ICU admission, necrotizing or cavitary infiltrates, or empyema.

(https://academic.oup.com/cid/article/52/3/e18/306145 Last Accessed: August 27, 2024)

Level of Evidence: A-III (Good supporting evidence from opinions of respected authorities, based on clinical experience, descriptive studies, or reports of expert committees)

For adults who present with presumed viral CAP, it is unclear whether antibiotic treatment is beneficial. However, when the patient with CAP has epidemiologic, clinical, or laboratory evidence of active influenza, a neuraminidase inhibitor should be included in the treatment regimen [32,235]. During the COVID-19 pandemic, patients with upper respiratory tract infection progressing to clinical and/or radiographic signs suspicious for viral pneumonia should immediately receive SARS-CoV-2 diagnostic testing and consideration for COVID-19 treatment protocol. Updated guidance for managing patients with COVID-19 pneumonia is available at the IDSA website [94].

The time to the first dose of antibiotics for adults with CAP has engendered debate. A 2003 guideline developed by the IDSA recommended initiation of antibiotic therapy within four hours after hospitalization. Quality measures linked to this timeframe were developed by the Joint Commission and the Centers for Medicare and Medicaid Services [2,66,79,80]. Experts have criticized the timeframe requirement, with some noting that it has the potential to result in less-than-optimal care and others adding that diagnosis of pneumonia in the emergency department is challenging, especially in older patients who have an atypical presentation [51,52,79,80]. In a survey of 121 emergency physicians, 55% of the respondents said they had prescribed antibiotics to patients they did not believe had pneumonia in an effort to comply with the Centers for Medicare and Medicaid Services quality measure; 42% of these respondents said they had prescribed as such more than three times a month [80]. Sixty percent of the respondents said they did not believe that the guideline improves patient care. The results of a systematic review and a large-scale study have shown no decrease in mortality with a first dose administered within four hours [57,81,82].

As emphasized by the IDSA/ATS guideline committee, the recommendation at present is to begin antibiotic treatment promptly, without delay, administering the initial dose at the site of care (e.g., emergency department, clinic, office) where the diagnosis is first made [235].

With the availability of well-absorbed, effective oral antibiotics, hospitalized adults do not require intravenous antibiotics for the duration of treatment. Intravenous therapy can be changed to an oral regimen when the patient is hemodynamically stable, improving clinically, and able to take oral medications safely [235]. For patients on a general ward floor, this transition can often be made by the third hospital day; patients in the ICU usually reach this point within seven days. It is recommended that the oral antibiotic be either the same drug or within the same drug class as the intravenous antibiotic. Patients can be discharged from the hospital as soon as clinical stability has been achieved, provided they have no comorbidities requiring inpatient care and have a safe home environment and reliable follow-up. The IDSA/ATS note the following criteria for determining clinical stability [235]:

The IDSA/ATS recommend that antibiotic therapy be given for a total of at least five days. The duration of therapy should be extended at least 48 to 72 hours beyond resolution of fever, assuming significant clinical improvement and no more than one pneumonia-associated active clinical sign [235]. A five- to seven-day course should suffice for most uncomplicated cases that show a prompt and satisfactory response to treatment.

The duration of treatment for gram-negative bacillary and staphylococcal pneumonia bears further comment. Unlike pneumococcal pulmonary infection, which usually heals without residual damage, these pathogens often cause destructive changes and small cavities in the lung, which clear slowly and heal by fibrosis. Thus, a more prolonged course of therapy (two to three weeks) should be considered, depending on severity of illness and response to therapy.

The clinical response to initial antibiotic therapy is unsatisfactory in approximately 15% of adults with CAP [47]. Failure to respond has no clear definition, and the IDSA/ATS guideline suggests using a systematic classification of cases, with attention to timing and character of response, as a guide to further evaluation and management. In general, treatment failures may be classified as persistent or non-responding, as a delay in achieving clinical stability, or as progressive pneumonia with clinical deterioration. Some clinical deterioration during therapy is not uncommon in the first 24 hours of treatment; as many as 45% of adults admitted to the hospital later require transfer to the ICU [47]. When the diagnosis of CAP is correct and guideline-recommended therapy has been used, the most common reason for treatment failure is an inadequate host response. For these patients, the appropriate management depends on individual case considerations, such as comorbidities, adequacy of pulmonary toilet, and whether the intravenous regimen has been reliably and consistently administered.

Guideline-directed management of CAP has been associated with many benefits. In one study, use of guideline-recommended antibiotics was associated with a significantly shorter time to clinical stability; clinical stability was achieved by seven days in 71% of patients treated with guideline-recommended antibiotics and in 57% of those treated with nonadherent regimens [15]. Adherence to recommendations guiding the selection of antibiotics was also associated with a significantly shorter length of stay (8 vs. 10 days) and a significantly lower overall in-hospital mortality rate (8% vs. 17%) [15]. In a Canadian study of adults (mean age: 51 years) who, in the main, had mild pneumonia, guideline-adherent selection of antibiotic treatment was associated with a lower mortality rate (1%) than that found when treatment selection that was not adherent to guidelines (6%) [83]. The mortality rate associated with the use of macrolides was also significantly lower than that with the use of fluoroquinolones (0.2% vs. 3%) [83]. In a large study of 54,619 patients who were hospitalized at 113 community hospitals (not in the ICU), use of guideline-adherent treatment was associated with a lower in-hospital mortality rate, lower rate of sepsis and renal failure, and shorter length of stay and duration of parenteral therapy [17]. Decreased mortality has also carried over to populations with more severe disease, with nonadherent therapy being associated with an increase in inpatient mortality (25% vs. 11%) among older adults (median age: 71 years) who were admitted to an ICU [16]. In addition to the higher rates of adverse outcomes, the low rate of adherence has also resulted in the inappropriate use of antimicrobials in at least half of cases [21].

Despite the benefits of guideline-directed treatment and the wide dissemination of the guidelines for management of pneumonia in adults, adherence has been low, especially with regard to antibiotic selection, with rates ranging from 9% to 82% [15,16,17,18,20]. In a study of more than 34,000 patients in a managed care organization, adherence to the 2003 IDSA guidelines in ambulatory settings was 52% for patients who were previously healthy and had not had recent exposure to antibiotics [20]. The rate of adherence was better (82%) for patients who had comorbidities and no recent exposure to antibiotics [20]. One study found that most cases of guideline-discordant use of antibiotics for older adults represent undertreatment [15]. The use of recommended antibiotics in the emergency department significantly increased from 1993 through 2008, but the percentage of patients receiving these drugs is still not optimal, with 60% to 70% of patients not receiving recommended antibiotics [84].

As the low rate of guideline adherence demonstrates, disseminating clinical practice guidelines alone is not enough to change practice. Physician education should address barriers to guideline adherence, including lack of familiarity, concerns about the practicality of recommended antibiotics, increased cost, lack of documented improved outcomes, and potential conflict with other guidelines [23]. Physician practices and healthcare systems should implement strategies that have changed physician behavior in other health condition settings, such as face-to-face educational outreach, use of local opinion leaders, and individualized audit with peer-comparison feedback [85]. In a study of six Dutch hospitals, significant increases in adherence to guideline-recommended care were achieved with an intervention that included the establishment of a local committee, a lecture by a respected opinion leader, feedback on performance, and critical care pathway pocket cards [86]. The intervention also included a second phase that focused on aspects of treatment in most need of improvement. In another study, weekly e-mail reminders listing performance data on antibiotic administration recommendation for individual emergency physicians helped to increase guideline adherence [87]. The use of a standardized evidence-based order set was associated with a decrease in mortality and was also cost-effective [88].

Evidence suggests that severe pneumonia is a cause of long-term morbidity and excess mortality among adults. In a population-based follow-up study of adults with CAP in Canada, conducted over a median of four years, the re-hospitalization rate for pneumonia was 16% to 72% for all causes [9].

The PSI classification and the time to clinical stability can both help predict adverse outcomes. Mortality has been reported to be higher for people originally classified as PSI class V than PSI classes I and II, with rates of 82% compared with 15% [9]. A time to clinical stability of more than 72 hours has been associated with a significantly higher rate of adverse outcomes than shorter times [90]. Overall, severe CAP has been associated with a 30-day re-hospitalization rate as high as 20%, a 30-day mortality rate as high as 23%, and all-cause mortality within one year as high as 28% [76].

These findings indicate that adults with severe pneumonia should be followed up closely to monitor for adverse events after discharge. The time to clinical stability is a useful guide for a follow-up plan; patients in whom clinical stability is not achieved until more than 72 hours after admission should be seen in follow-up soon after discharge [3,90]. Strategies to prevent influenza and pneumonia should also be emphasized for all hospitalized patients. When indicated, immunization against pneumococcal infection should be initiated before or shortly after discharge, as recommended by the Advisory Committee on Immunization Practices (ACIP) and others [47,91,227].

After a diagnosis of pneumonia has been made, patient education should include directions for use of the antibiotic and information on potential untoward effects of the drug. Follow-up instructions, depending on the clinical situation, may include 24-hour telephone contact or follow-up in the office after 24 to 48 hours. This will improve adherence to the prescribed therapy, provide an opportunity to address side effects of drug therapy, and allow progress to be monitored. The need for hospitalization should be assessed throughout the course of the illness. Education should also include instructions to drink plenty of fluids and to use an antipyretic to control fever and myalgias when needed. Use of cough suppressants should be avoided, as the cough reflex and sputum expectoration enhance removal of thick secretions. However, in the event of a constant, nonproductive cough, as found especially with mycoplasmal infection, a narcotic such as codeine at night may allow for more restorative sleep.

Provisions for patients with limited English language proficiency are required under federal law, and the U.S. Department of Health and Human Services and the Office of Civil Rights view a lack of adequate interpretation as discrimination, based on the Civil Rights Act of 1964 [19]. According to U.S. Census Bureau data, nearly 68 million Americans speak a language other than English at home, with more than 25 million (8.2% of the population) reporting that they speak English less than "very well" [127]. Immigrant patients with chronic illness may feel unable to return to their home countries due to a lack of available medical care. Changes in healthcare law restricting federal funding of services to only legal residents may cause significant problems for certain facilities, with conflicts arising from providing life-saving care for patients who have no means of reimbursement and no medical services waiting for them in their home countries.

Viral pathogens are reported to be responsible for most cases of CAP in preschool-aged children and as many as 80% of cases in children younger than 2 years of age [30]. In children younger than 2 years of age, the most common viral pathogen, occurring in up to 40% of cases, is RSV; other viral pathogens include adenoviruses, bocavirus, human metapneumovirus, influenza A and B viruses, parainfluenza viruses, coronaviruses, and rhinovirus [9,29,30,32].

RSV infection is common in infants and young children; it is estimated that most children have had RSV by 2 years of age [31]. It is the leading cause of pneumonia in infants younger than 1 year of age, with 25% to 40% of those infected developing signs of pneumonia or bronchiolitis [29]. Premature birth, very young age, compromised immune system, and impaired lung or heart function are all risk factors for RSV-related pneumonia in infants. In contrast to preschool-aged children, the percentage of viral cases is much lower among older children and adolescents (10 to 16 years of age), and pneumonia caused by RSV is rare in this population.

In older children, viral and atypical bacterial infection account for most mild CAP managed in the ambulatory setting, while pyogenic respiratory bacterial infection is responsible for the majority of CAP in seriously ill, hospitalized children [30]. S. pneumoniae is the most common bacterial pathogen in school-aged children. Studies show that atypical pathogens account for 3% to 23% of cases, most commonly mycoplasma in older children and chlamydia in infants and young children [30]. A 2009 European study examining causative agents in hospitalized pediatric patients with radiographic evidence of pneumonia found bacterial infection in 53% of patients and viral pathogens in 67% of patients, with 33% of children in the study showing evidence of both [63]. S. pneumoniae was the most common bacterial pathogen (46%), followed by M. pneumoniae and C. pneumoniae. The primary viral pathogens identified were influenza A or B, parainfluenza, rhinovirus, RSV and, human metapneumovirus [63].

As with adults, severe CAP caused by S. aureus is encountered during outbreaks of influenza [223]. Legionella spp. and fungal pathogens are uncommon in children. A combination of viral and bacterial pathogens occurs in up to half of children with CAP [30,32].

The clinical presentation of CAP in children is similar to that in adults, but can vary according to age and developmental stage. For example, cough productive of purulent sputum may be elicited in older children, but nonproductive cough is common in young children and infants [30,60]. Nonspecific irritability and restlessness may be the primary symptoms in infants.

During the physical examination of pediatric patients, the clinician should look for signs of hypoxia and dehydration, as well as retractions, tachypnea, and use of accessory muscles of respiration [60]. The clinician should also evaluate the upper respiratory tract for evidence of rhinorrhea, otitis media, and pharyngitis [60]. Auscultation of the chest should be carried out, and the Pediatric Infectious Diseases Society (PIDS)/IDSA guideline recommends pulse oximetry for children with suspected hypoxemia [30].

One of the most common reasons for pediatric emergency room visits is fever, and fever is present in 88% to 96% of identified pneumonia cases in developed countries [70]. However, children with fever and wheezing commonly have either upper respiratory disease or reactive airway disease. As with pneumonia in adults, the accuracy of any one sign or symptom in predicting the likelihood of pneumonia is limited [61]. Nonspecific symptoms such as vomiting and abdominal discomfort are common. Careful attention should be given to the chest exam, as diminished breath sounds and fine end-inspiratory crackles are subtle, important clues to the presence of pneumonia in the pediatric patient. In one study, non-specific crackles were present in more than 90% of children with pneumococcal or mycoplasma pneumonia [70]. Infants with pneumonia commonly present with poor feeding and irritability as well as tachypnea, retractions, grunting, and hypoxemia; cough is rare [64].

Several clinical rules have been developed for predicting the likelihood of pneumonia in children on the basis of discernible clinical signs. The presence of at least two of the following signs—fever, tachypnea, and reduced oxygen saturation—is associated with a high probability of the disease; the absence of all three indicates a low probability [61]. Other signs of respiratory distress, such as cough, nasal flaring (in infants), rales, and decreased breath sounds, have also been found to be independent predictors of pneumonia in infants and children [60,62]. Bronchial breath sounds, rales, and dullness to percussion are more likely to occur in older children and adolescents [60].

Unlike diagnosis in adults, a chest radiograph is not the diagnostic standard to be applied for all CAP in children. The PIDS/IDSA guideline notes that routine chest radiographs are not necessary for children who can be treated as outpatients [30]. However, posteroanterior and lateral chest radiographs should be obtained when there is fever and respiratory distress suspected or documented hypoxemia, or illness severe enough to warrant hospitalization [30]. In a study of 99 children hospitalized with what was later determined to be pneumonia, the most common abnormal finding was "diminished" breath sounds; only 21% were described as having "normal" breath sounds. Radiographic evidence of pulmonary consolidation was present in 79% of patients, and correlation between diminished breath sounds and a positive chest x-ray was 60.2% [63].

Laboratory Tests

Unlike the situation in adults, titers of shed virus in children are high [31]. Thus, rapid antigen testing of nasal or throat swabs for influenza and other respiratory viruses should be done for infants and young children [30]. However, it should be noted that negative results of influenza virus on rapid antigen tests do not conclusively rule out infection with influenza virus. Testing for C. pneumoniae is not recommended.

Blood cultures are not routinely needed but should be obtained in children hospitalized for moderate-to-severe pneumonia that is presumed to be bacterial [30]. Urinary antigen detection tests often have false-positive results in children and are therefore not recommended for the diagnosis of pneumococcal pneumonia.

The PIDS/IDSA guideline, published in 2011, addresses the management of CAP in children 3 months of age and older who are otherwise healthy; the guideline does not provide guidance for neonates and infants younger than 3 months of age or children with comorbidities [30]. The guidelines were developed in an effort to decrease morbidity and mortality, as had been shown with the guideline for adults. Similar to the IDSA/ATS guideline, the management issues addressed in the PIDS/IDSA guidelines are site of care and selection and duration of antibiotic therapy, as well as adjunctive surgical and nonantibiotic treatment for complications. As with the guideline for adults, treatment of pneumonia-related symptoms is not included in the pediatric guideline. The discussion here is limited to site of care and antibiotic therapy.

Site of Care

To aid in making site-of-care decisions, the PIDS/IDSA guidelines recommend that a child or infant with CAP be hospitalized if any of the following factors are present [30]:

Most children with pneumonia do not require care in an ICU. The guideline states that a child should be admitted to an ICU or a unit with continuous cardiorespiratory monitoring capabilities if the child [30]:

Selection of Antibiotics or Antivirals

The American Thoracic Society and the IDSA recommend that anti-influenza treatment, such as oseltamivir, be prescribed for adults with CAP who test positive for influenza in the inpatient setting, independent of duration of illness before diagnosis.

(https://www.atsjournals.org/doi/full/10.1164/rccm.201908-1581ST Last Accessed: August 29, 2024)

Strength of Recommendation/Level of Evidence: Strong recommendation; moderate-quality evidence

The PIDS/IDSA guideline recommends empiric antibiotic therapy according to patient age, immunization status, and site of care. Among infants and children 3 months to 5 years of age, antibiotic therapy is not routinely recommended because viral infection is the predominate cause of CAP in this age group [30]. When influenza virus is the suspected cause of pneumonia, influenza antiviral therapy should be started as soon as possible, as maximal benefit is achieved when treatment begins within 48 hours of the onset of symptoms. (Treatment should not be delayed while waiting for the results of viral testing.) The PIDS/IDSA guideline recommends three U.S. Food and Drug Administration (FDA)-approved influenza antiviral therapies: oseltamivir (Tamiflu), zanamivir (Relenza), and amantadine (Symmetrel) [30]. A fourth antiviral therapy, rimantadine (Flumadine), is included in the guideline, with a note that the agent is FDA-approved for prophylaxis—not treatment—in children 1 year of age and older [30]. The guideline adds that data on the safety and efficacy of the agent for children 1 year of age and older have been published.

As in adults, S. pneumoniae is the most common bacterial cause of CAP in children; thus, if a bacterial pathogen is thought to be the cause, amoxicillin or amoxicillin/clavulanate is recommended as first-line therapy for mild-to-moderate illness in previously healthy children 3 months to 5 years of age who are up-to-date with immunization [30]. Several alternatives can be used for children who are allergic to amoxicillin (Table 8). Amoxicillin is also the preferred antibiotic for mild-to-moderate CAP in adolescents and children 5 years of age and older [30]. For children of all ages, especially children older than 5 years of age and adolescents, a macrolide is recommended if an atypical bacterial pathogen is thought (or documented) to be the cause.

For fully immunized infants and school-aged children who are hospitalized, treatment with ampicillin or penicillin G is recommended when local epidemiologic data show a low level of penicillin resistance to S. pneumoniae [30]. For children who are not fully immunized or are hospitalized in an area with a high level of penicillin-resistant S. pneumoniae, treatment with a third-generation cephalosporin (ceftriaxone or cefotaxime) should be given intravenously. If M. pneumoniae or C. pneumoniae is strongly suspected, treatment should include a macrolide (orally or intravenously) with a ß-lactam and diagnostic testing should be done as soon as possible [30]. The PIDS/IDSA guideline also recommends antimicrobial treatment for specific pathogens; however, a discussion of all possible pathogens is beyond the scope of this course.

According to a systematic review, zinc supplementation in addition to standard antibiotic therapy was not shown to have significant benefit on clinical recovery of severe or nonsevere pneumonia in children 2 to 59 months of age [89].

Duration of Therapy

Most studies have evaluated 10-day therapy, and this duration is associated with good outcomes. However, a shorter duration may be equally as effective, especially for mild disease treated on an outpatient basis [30].

Benefits of Guideline Adherence

Because the PIDS/IDSA guideline for management of CAP in children is relatively recent, data are lacking on the benefits of guideline-adherent treatment in the pediatric population. One study did show that more children received appropriate antibiotics after the development of a clinical practice guideline based on the PIDS/IDSA guideline and an antimicrobial stewardship program [14]. It is assumed that more data will become available over time.

Late Complications

Data on the long-term effects of pneumonia during childhood are lacking. A systematic review demonstrated that severe pneumonia in children younger than 5 years of age is associated with long-term sequelae, with restrictive lung disease being the most common sequela [95]. Overall, major respiratory sequelae (e.g., restrictive lung disease, obstructive lung disease, bronchiectasis) occurred in 5.5% of children treated on an outpatient basis and in 13.6% of children hospitalized for treatment [95]. Sequelae occurred in approximately 54% of children who had pneumonia caused by adenovirus.

Pneumococcal vaccines have been improved over time by broadening the coverage of serotypes in the vaccine to include those that are causing the most common invasive infections. In the past, a single agent, the 23-valent polysaccharide vaccine (PPSV23) (Pneumovax), had been recommended for use in selected adults with conditions of impaired immunity, and for all adults older than 65 years of age [96]. This vaccine provided some protection against 85% to 90% of the pneumococcal serotypes that cause invasive disease in these populations [97]. In 2021, PCV15 (replacing PCV13) and PCV20 were introduced for adults [240].

The use of pneumococcal conjugate vaccines in the pediatric age group has been followed by a reduction in the incidence of pneumococcal disease among children, and, indirectly, among adults as well. By 2013, the incidence of invasive pneumococcal disease caused by serotypes represented in the PCV13 vaccine had declined in the adult population older than 65 years of age by approximately 50% compared with 2010 [227]. In 2012, upon approval by the FDA, the ACIP recommended the use of PCV13 for adults with immune deficits and other conditions that impose a heightened risk for invasive pneumococcal infection. After reviewing additional data in 2014, the ACIP extended its recommendation for PCV13 use to all adults older than 65 years age [227].

In 2021, the ACIP again amended its recommendation for PCV use in older adults, based on sharp declines in pneumococcal disease among adults since the advent of PCV13 use in children [227]. The ACIP now recommends a routine single dose of PCV20 for adults older than 65 years of age (Table 10). Alternatively, one dose of PCV15 may be administered followed by PPSV23 given at least one year after the PCV15 dose. A minimum interval of eight weeks between PCV15 and PPSV23 can be considered for adults with an immunocompromising condition, cochlear implant, or cerebrospinal fluid (CSF) leak to minimize the risk of invasive pneumococcal disease caused by serotypes unique to PPSV23 in these vulnerable groups [96,227,240].

As of June 2024, ACIP recommends PCV21 as an option for adults aged 19 years of age or older who currently have a recommendation to receive a dose of PCV [239]. Current information, schedules, and guidance for adult immunizations is maintained at the CDC/ACIP website at https://www.cdc.gov/vaccines/schedules.

The influenza vaccine is developed each year to contain the three virus strains that are expected in the upcoming influenza season. The vaccine has traditionally been a trivalent inactivated vaccine (TIV), but in 2003, a trivalent live, attenuated influenza vaccine (LAIV) was introduced in the United States [97]. In 2010, a new high-dose formulation of TIV became available. The LAIV, which contains four times the amount of influenza antigens as other TIVs, is designed to induce a higher immune response in older people [97]. The LAIV is administered as a nasal spray.

The ACIP once recommended a risk-stratified approach to influenza vaccination, but it updated its recommendations to universal vaccination beginning in the 2010–2011 influenza season (Table 10) [91]. Routine annual influenza vaccination is recommended for all persons 6 months of age and older who do not have contraindications. The ACIP provides updated recommendations for the use of licensed and age-appropriate seasonal influenza vaccines in the United States. Current ACIP influenza vaccine recommendations are available at the CDC website [236].

In their guideline for the management of CAP, the IDSA/ATS make the following strong recommendations for prevention based on the ACIP recommendations [47]:

The IDSA/ATS recommends that vaccination status be assessed in all patients with CAP at the time of hospital admission, especially in those with underlying medical conditions [47]. If vaccination is needed, it may be done either at hospital discharge or during outpatient treatment.

The PIDS and the IDSA also echo the ACIP recommendations in their guideline [30]:

Declining rates of pneumonia and pneumonia-related deaths are thought to represent the effectiveness of influenza and pneumococcal vaccination [40,98,99]. In a study of a community-dwelling older population, influenza vaccination decreased the risk of hospitalization for pneumonia or influenza, as well as the risk of death, across 10 influenza seasons [7]. Systematic reviews and meta-analyses have shown that pneumococcal vaccination reduces the incidence of invasive pneumococcal disease in both older adults and children, although the findings are unclear for adults with chronic illness [100,101]. Other studies of adults have shown that pneumococcal vaccination is associated with benefit in terms of a lower risk of adverse outcomes associated with the disease. For example, in a study of nearly 3,500 older people (median age: 75 years) who were hospitalized for CAP, the rate of mortality or ICU admission was 40% lower among those who had received prior PPSV23 vaccination [8].

Among children, the introduction of the PCV7, PCV13, and later PCV15 has led to a substantial decrease in the rate of invasive pneumococcal disease, but the decrease in the rate of CAP has been less dramatic. Early studies showed substantial improvements in the hospitalization rate for CAP only among young children. In one study, the hospitalization rate decreased 39% for children younger than 2 years of age [98]. In another study, the decrease was substantial only for children younger than 1 year of age (22%) and was minimal for children 1 to 5 years of age; the rate increased for adolescents and children older than 5 years of age [4]. The rate of outpatient CAP visits has not changed significantly for this population [5,10].

Despite the wide distribution of the ACIP immunization schedule and public campaigns about the importance of vaccination, rates of both pneumococcal and influenza vaccination remain relatively low. According to CDC estimates, influenza vaccination coverage for the 2021–2022 season among adults 18 years of age or older was 49.4%, a decrease of 0.8 percentage points from the prior season [102]. Coverage among children 6 months through 17 years of age was 57.8%. Roughly half (51.4%) of all persons 6 months and older were vaccinated during the 2021–2022 season. Influenza vaccination coverage was lower among Hispanic (37.9%) and non-Hispanic Black (42.0%) adults than White (53.9%) adults [102]. Coverage among adults 65 years of age or older (73.9%) was higher compared with younger age groups (37.1% to 52.4%). Among healthcare personnel, influenza vaccine coverage during the 2021–2022 influenza season was 79.9%. When analyzed by setting, vaccine coverage was highest among healthcare personnel in ambulatory care and hospital settings with vaccine requirements (81.2% and 92.0%, respectively), and lowest in medical settings without vaccination requirements or recommendations (48.1%) [103].

The national rate of influenza vaccination among all adults has improved over the past decade, yet racial disparities persist. Comparing rates at five-year intervals from 2005 to 2015, the rate of vaccination more than doubled for adults younger than 65 years of age in all ethnic categories (Table 11) [113]. The rate disparity between white adults (44.2%) and that observed for black (36.7%) and Hispanics (31.2%) remains evident. Previous studies have also shown higher rates of vaccination for white older adults compared with black and Hispanic older adults [104,105,106,107]. Racial disparities have also been found when rates of pneumococcal and influenza vaccination for residents of long-term care facilities were compared, with substantially lower rates for black residents [108,109,110].

In 2018, the estimated overall rate of pneumococcal vaccination coverage among adults older than 65 years of age was 69% [237]. The rate was substantially lower (approximately 23%) among younger adults in high-risk groups. Selected data from this report are summarized in Table 12[237].

In addition, adherence to recommendations for pneumococcal and influenza vaccination of older adults admitted to the hospital has been low. In a study of nearly 105,000 patients 65 years of age and older who had not received either vaccination before admission to the hospital, 99.4% did not receive the pneumococcal vaccine and 97.3% did not receive the influenza vaccine before hospital discharge [111].

Rates of both pneumococcal and influenza vaccination are higher among children than adults. Overall, approximately 83% of children 19 to 35 months of age have received at least four PCV13 doses [112]. The rate varies according to race/ethnicity, with the lowest rates among Asian and black children (Table 13) [112].

In its Healthy People 2030 initiative, the U.S. Department of Health and Human Services has set objectives for improving vaccination rates among adults and children [114]. To reach these targets, healthcare providers must address documented barriers to recommended vaccinations and gain a better understanding of other challenges to vaccination. Unequal access to health care appears to account for a low percent of racial disparities [105]. Rather, lack of awareness of the need for vaccination and misconceptions about vaccines have been reported as the primary barriers in several studies [104,105,106,115,116,117].

Among adults, misconceptions about vaccines range from the belief that healthy people do not need vaccinations to a fear of side effects [104,106,116]. Beliefs about vaccines vary by race/ethnicity, age, education, and gender. For example, in a survey of more than 6,700 older adults, lack of awareness that influenza vaccination was needed was more common among Hispanic (33%) and black individuals (25%) than among white individuals (21%) [105]. In contrast, concern about side effects was more common among white individuals (15%) than among black and Hispanic individuals (10% and 6%, respectively) [105]. The belief that vaccination would not prevent illness was consistent across the racial/ethnic groups. In other studies, lower rates of influenza vaccination among older black adults have been significantly associated with lower rates of positive attitudes about vaccination [105,118]. It is unclear whether the negative attitude represents mistrust of the vaccine itself or of healthcare/healthcare providers in general [105]. The findings of one study showed that, compared with white adults, more black and Hispanic adults believed that they had become sick from a previous influenza vaccination [106]. Language proficiency and level of acculturation have been associated with lower vaccination rates among older Hispanic adults [107,119].

Parental attitudes about vaccines are an important factor in vaccination rates among children. The primary attitude is concern about the safety and efficacy of the vaccine, including fear of adverse events, the discomfort associated with vaccination, distrust of advocates of vaccination, and belief that the vaccine should not be given when a child has a minor illness [117,120,121,122]. Difficulty remembering or confusion about the vaccination schedule for children is also a major challenge [120,122]. Changes in access to health care have been noted as a factor in the low rate of influenza vaccination among teenagers [117].

Healthcare provider-related factors should also be addressed. Slightly more than half of older adults have said that their healthcare provider did not recommend influenza vaccination, and this percentage has been consistent across races/ethnicities [105,106]. The lack of provider recommendation may be a misperception or may be a reality. It has been noted that nearly half of providers do not follow the ACIP recommendations for vaccination [116]. Provider recommendation is essential, as it has been found to be the strongest predictor of whether a person will receive vaccination, even among those who have negative attitudes toward vaccines [104,106,115,116,123]. Providers have said that the lack of an effective reminder system is a factor in low vaccination rates [116,123].

Strategies to improve rates of vaccination and other preventive measures rely on effective patient-clinician communication. Among the most important factors for effective communication across all healthcare settings are knowledge of the language preference of the patient and family; an awareness of the patient's and family's health literacy levels; and an understanding of and respect for the patient's and family's cultural values, beliefs, and practices [124,125,126]. These issues are significant, given the growing percentages of racial/ethnic populations. According to U.S. Census Bureau data from 2013, more than 60.3 million Americans speak a language other than English in the home, with more than 25.1 million of them (8.6% of the population) reporting that they speak English less than "very well" [127]. Clinicians should ask their patients what language is spoken at home and what language they prefer for their medical care information, as some patients prefer their native language even though they have said they can understand and discuss medical information in English [128]. When the healthcare professional and the patient speak different languages, a professional interpreter should be used. Studies have demonstrated that the use of professional interpreters rather than "ad hoc" interpreters (e.g., untrained staff members, family members, friends) facilitates a broader understanding, leads to better outcomes, and is better aligned with patient preferences [129,130,131].

Studies have indicated that as many as 26% of Americans have inadequate health literacy, which means they lack the ability to understand health information and make informed health decisions; an additional 20% have marginal health literacy [132,133,134]. Health literacy varies widely according to race/ethnicity, level of education, and gender. Clinicians are often unaware of the literacy level of their patients and family, but several instruments are available to test the health literacy level [126,135]. These instruments vary in the amount of time needed to administer and the reliability in identifying low literacy. Among the most recent tools is the Newest Vital Sign (NVS), an instrument named to promote the assessment of health literacy as part of the overall routine patient evaluation [136]. The NVS takes fewer than three minutes to administer, has correlated well with more extensive literacy tests, and has performed moderately well at identifying limited literacy [126,135]. Two questions have also been found to perform moderately well in identifying patients with inadequate or marginal literacy: "How confident are you in filling out medical forms by yourself?" and "How often do you have someone help you read health information?" [126]. Clinicians should adapt their discussions and educational resources to the patient's and family's identified health literacy level and degree of language proficiency and should also provide culturally appropriate and translated educational materials when possible.

Cultural competency is essential for addressing healthcare disparities among minority groups [124]. Clinicians should ask the patient about his or her cultural beliefs, especially those related to health, and should be sensitive to those beliefs.

Targeted evidence-based strategies can help clinicians improve vaccination rates (Table 14). Education about the importance of vaccination is the cornerstone of most strategies. Messages should be clear and emphasize the benefits of vaccination and the risks of not receiving vaccination. Acknowledging the risks of vaccines can help enhance patient trust [117]. Clinicians should give their patients a list of online resources that provide balanced information on vaccines (Table 15). Differences in beliefs about vaccines across racial/ethnic groups indicate that targeted messages developed for specific demographic subgroups may be useful [219]. In addition, language-specific educational resources may also help increase vaccination rates by enabling patients to better understand the need for vaccination and its safety.

Education and provider recommendation are particularly important for people at high-risk for severe health impacts of pneumonia, as the lowest vaccination rates are reported for these patients [102,103]. One survey showed that provider recommendations for pneumococcal and influenza vaccination were low for this population; the rate of recommendation was lowest for people with a weakened immune system and those receiving radiation therapy or chemotherapy (Table 16) [116]. Clinicians should identify high-risk patients in their practice and take special steps to ensure that these patients receive appropriate vaccinations.

Missed opportunities represent another practice-related area in which clinicians can improve vaccination rates. Although many clinicians check immunization status during well visits, most do not check the status during acute visits, nor do they take advantage of the visit to administer the vaccination [105,115]. Healthcare providers can close the gap on missed opportunities for vaccination by taking advantage of every office visit to administer vaccinations, reminding their patients about the need for vaccination, or scheduling a future appointment for vaccination [105,115,117]. Educational fliers and pamphlets in the waiting room and examination rooms can engage patients and parents and help prompt discussions about vaccination [116].

Patient reminder and recall systems in primary care settings have been effective in improving vaccination rates. A meta-analysis found that rates among both children and adults increased up to 20% with several types of reminders, including postcards, letters, and phone calls [137]. The most effective reminder system was phone calls, but it was also the most expensive. Given that about 25% of primary care physicians currently use reminder systems, increasing the number of physicians who use such systems can in turn increase vaccination rates [123]. Standing orders for vaccinations have been shown to substantially increase vaccination rates, yet are used by only 20% to 33% of physicians [123,138]. Again, adopting this system results in improved vaccination rates.

Many people have turned to facilities outside of their primary healthcare provider to receive vaccinations. Health fairs, pharmacies, grocery stores, senior centers, and workplaces have become more common settings for vaccination because of their convenience and lower cost [123,138]. Clinicians can also help increase vaccination rates by participating in community events that provide vaccinations and by promoting these settings as alternative options.

Programs to provide vaccinations to high-risk patients in the emergency room have been successful at increasing vaccination rates [139,140]. In a three-week intervention program at one inner city emergency department, participants were provided appropriate immunizations when they were at high risk for specific diseases [139]. During the study period, rates of influenza and pneumococcal vaccinations increased from 16% to 83% and from 18% to 84%, respectively. Such programs can help healthcare systems adhere to guideline recommendations for vaccinating hospitalized patients.

In a systematic review, the American College of Physicians found several patient-related and surgery-related factors that increased the risk of postoperative pulmonary complications. The most common patient-related factors were the presence of COPD and an age older than 60 years [145]. Other significant factors were an American Society of Anesthesiologists (ASA) class of 2 (defined as a patient with mild systemic disease) or higher, functional dependence, and congestive heart failure. Cigarette use was associated with a modest increase in risk, and obesity and mild or moderate asthma were not found to increase risk [145]. Use of a PPI or histamine2 receptor antagonist is also thought to be a risk factor [45]. Surgery-related factors included prolonged duration of surgery (i.e., more than three to four hours), emergency surgery, and surgical site, with abdominal surgery, thoracic surgery, neurosurgery, head and neck surgery, vascular surgery, and aortic aneurysm repair being associated with the greatest risks [145].

The risk for VAP appears to be greatest during the first week after intubation. In one study, the risk was estimated to be 3% per day during the five-day period following intubation, decreasing to 2% per day for days 5 through 10, and to 1% per day for longer durations [147]. In a population of children who had cardiothoracic surgery, pneumonia risk correlated with mechanical ventilation for longer than three days [144]. Nearly half of all cases of VAP develop within the first four days of mechanical ventilation [148].

Other identified risk factors among adults include prolonged placement of the patient's head in the supine position; use of a nasogastric tube, paralytic agents, or PPI or histamine2 receptor antagonist; advanced age; chronic lung disease; and head trauma [45,149]. Among children, VAP has been significantly associated with subglottic/tracheal stenosis, trauma, and tracheostomy [150]. In one study, VAP was most frequently associated with ICU admission diagnoses of postoperative care, neurologic conditions, sepsis, and cardiac complications [151].

The risk factors reported to be associated with nursing home-acquired pneumonia include profound disability, immobility, urinary incontinence, deteriorating health status, difficulty swallowing, and inability to take oral medications [42]. Older age, male gender, and antipsychotic and anticholinergic medications have also been reported to increase risk [23,42].

Among adults with no previous antibiotic exposure, the most common bacterial causes of HAP are S. pneumoniae, H. influenzae, Escherichia coli, K. pneumoniae, and S. aureus [26,28,35,148]. Gram-negative bacilli resistant to first-generation cephalosporins also frequently develop in late-onset HAP. For up to 40% of adults with previous antibiotic exposure, late-onset HAP is caused by potentially multidrug-resistant pathogens, including Pseudomonas aeruginosa, Acinetobacter baumannii, and MRSA [26]. In a study of more than 3,600 patients admitted to an ICU, Pseudomonas spp. was the cause of pneumonia in 25% of patients; MRSA in 18%; and Acinetobacter spp. in 6% [35]. Other studies have shown that S. aureus is common among patients who are in a coma or have diabetes or renal failure. P. aeruginosa is common among patients who have had a prolonged stay in the ICU, have received prior antibiotics or corticosteroids, or who have structural lung disease. Legionella is usually found in patients who have compromised immune systems [35].

The causes of HAP in children have not been well studied. However, outbreaks of pneumonia caused by RSV have been common in pediatric wards [28].

The most common pathogens associated with VAP in adults are S. aureus and P. aeruginosa, followed by Enterobacter spp., A. baumannii, and K. pneumoniae [26,148,152,153]. These bacteria are among those that have become resistant to antibiotics, and the frequency of infection with MRSA is increasing. Almost half of all cases are caused by infection with more than one pathogen [148]. Although bacteria are the primary causative agents, viruses and saprophytic fungi have also been implicated as well [154].

As with HAP, few data are available on the etiology of VAP in children. In one report, P. aeruginosa was the most common cause, accounting for 22% of cases [142].

The bacterial pathogens that cause pneumonia in residents of nursing homes (and other long-term care facilities) differ according to the severity of disease. S. pneumoniae and H. influenzae are the most common causes of mild-to-moderate pneumonia in long-term care facilities [155]. In cases requiring hospitalization, C. pneumoniae, S. aureus, and influenza virus are frequently observed as well. Patients with severe illness commonly are infected with methicillin-sensitive S. aureus or MRSA, gram-negative enteric pathogens, or P. aeruginosa [23,155].

For healthcare-associated or community-acquired MRSA pneumonia, the IDSA recommends IV vancomycin or linezolid 600 mg PO/IV twice daily, if the strain is susceptible, for 7 to 21 days, depending on the extent of infection.

(https://academic.oup.com/cid/article/52/3/e18/306145 Last Accessed: August 29, 2024)

Level of Evidence: A-II (Good evidence from one or more 1 well-designed clinical trial, without randomization to support a recommendation for use)

VAP is often caused by MRSA and gram-negative bacilli such as Acinetobacter spp. and Pseudomonas. Vancomycin has been considered the first choice for treatment of MRSA infections [154]. However, the ATS/IDSA guidelines note that linezolid may have advantages over vancomycin for pneumonia caused by MRSA [28]. Linezolid has been compared with vancomycin for the treatment of pneumonia caused by MRSA in many studies, and linezolid has been found to improve survival and to be more cost-effective [147,165,166,167,168]. In a 2008 study, the rate of early microbiologic cure was not significantly higher for linezolid than for vancomycin, although there were trends favoring linezolid in several secondary clinical outcomes, such as clinical cure; duration of ventilation, hospitalization, and stay in ICU; survival time not on a ventilator; and overall survival [169]. The findings led the authors to suggest that the benefit of linezolid may be related to factors other than bacterial clearance.

For cases of VAP caused by gram-negative bacilli that are susceptible only to aminoglycosides or polymyxins, the ATS/IDSA suggests both inhaled and systemic antibiotics, rather than systemic antibiotics alone [28]. It is also reasonable to consider adjunctive inhaled antibiotic treatment as a last resort for patients who are not responding to intravenous antibiotics alone, whether or not the infecting organism is multi-drug resistant.

According to a meta-analysis, a short fixed-course (7 or 8 days) of antibiotic therapy may be more appropriate than a prolonged course (10 to 15 days) for patients with VAP not caused by nonfermenting gram-negative bacilli [170]. The short course reduced recurrence of pneumonia caused by multiresistant organisms without adversely affecting other outcomes. Among patients with nonfermenting gram-negative bacilli, recurrence was greater after the short course.

The 2019 ATS/IDSA guideline for CAP provides a rationale for choice of antibiotic therapy without specifying distinct protocols for nursing home-acquired pneumonia. The ATS/IDSA recommended abandoning the HCAP categorization for purposes of management decision. Instead, emphasis is placed on local epidemiology and validated risk factors to determine need for empiric MRSA or gram-negative bacillary coverage, followed by de-escalation of treatment if cultures are negative [235]. As indicated, the consistently strong individual risk factors for respiratory infection with MRSA, P. aeruginosa, or other gram-negative bacilli are prior isolation of these organisms, and/or recent hospitalization and exposure to parenteral antibiotics.

For patients with ventilator-associated pneumonia, the American Thoracic Society and the IDSA recommend a seven-day course of antimicrobial therapy rather than a longer duration.

(https://www.thoracic.org/statements/resources/tb-opi/hap-vap-guidelines-2016.pdf Last Accessed: August 29, 2024)

Strength of Recommendation/Level of Evidence: Strong recommendation; moderate-quality evidence

The lack of adherence to guideline-directed treatment of pneumonia cases associated with healthcare facilities is evidenced by wide variations in practice. For example, one study showed that more than 100 different antibiotic regimens had been prescribed as initial treatment and that de-escalation therapy was used for only 22% of patients [151]. Adherence rates for pneumonia associated with healthcare facilities have been reported to be lower than rates of adherence to guidelines for treatment of CAP. In one survey, guideline-recommended antibiotics were used 78% of the time for CAP, compared with 9% for HCAP [18]. This lack of adherence was not due to unfamiliarity or disagreement with the guidelines; 71% of the survey respondents said they were aware of the guidelines, and 79% said they agreed with and practiced according to them. In contrast, another survey showed that fewer than half of physicians were familiar with the ATS/IDSA guideline for treatment of nursing home-associated pneumonia [23]. It is reasonable to expect that strategies used to enhance adherence to guidelines in the setting of CAP would also be beneficial in the setting of pneumonia associated with healthcare facilities. Thus, feedback on performance, reminder systems, standardized order sets, and education emphasizing outcomes and cost-effectiveness would be valuable.

The prevention of postoperative pneumonia has long been a part of initiatives to decrease complications among patients undergoing surgery. The Respiratory Risk Index was developed to classify patients as being at low, medium, or high risk for postoperative respiratory failure [26]. The factors in the index include the complexity of the surgery, the ASA status, and comorbidities.

Smoking triples the risk for pulmonary complications after surgery, and smoking cessation for at least eight weeks before surgery, when possible, is recommended for current smokers [26]. The risk for complications in patients with respiratory disease or congestive heart failure can be ameliorated by optimum treatment before surgery (e.g., treatment with steroids for patients with COPD or asthma) [26].

Effective pain management after surgery also helps to decrease the risk of pulmonary complications. For postoperative patients who are not mechanically intubated, the ability to cough and clear secretions is important for preventing pulmonary complications [26]. The use of incentive spirometry and deep breathing exercises are recommended, especially for people at high risk for pulmonary complications, as are frequent coughing and early movement (in bed and/or walking) [26,93,145]. Fair evidence supports the selective (rather than routine) use of a nasogastric tube after abdominal surgery [145].

Two guidelines were developed to focus specifically on the prevention of VAP; one was jointly developed by the Society for Healthcare Epidemiology of America (SHEA) and IDSA, and the other was jointly developed by the Canadian Critical Care Trials Group and the Canadian Critical Care Society [149,171]. In addition, the CDC guideline addresses the prevention of HAP and VAP [93]. All of these agencies suggest a multicomponent strategy for prevention of pneumonia. Compliance with guidelines, however, has been slow; nursing surveys demonstrate rates of adherence to specific preventive measures ranging from 15% to 50% [12,172]. Education is beneficial, and training sessions are a proven means to enhance knowledge and practice among healthcare professionals caring for intubated patients [173].

The Institute for Healthcare Improvement (IHI) found that implementation of its ventilator bundle, a collection of five prevention strategies drawn from these guidelines, led to a 45% reduction in the incidence of VAP [174]. The bundle includes the following interventions [174]:

The IHI how-to guide on preventing VAP provides several practical recommendations, and posting compliance with the ventilator bundle in a prominent place in the ICU can encourage and motivate staff (Table 18) [174].

Assessment of Readiness to Extubate

Because of the increasing risk of infection as the duration of ventilation increases, the primary goal is to extubate patients as early as possible. Thus, assessment of the readiness for extubation and weaning protocols are key aspects in the preventive approach [28,35]. Daily interruption of sedation until the patient is awake has been shown to significantly decrease the number of days on mechanical ventilation, from 7.3 days to 4.9 days in one study [175]. There are risks to this approach, including the potential for increased pain, anxiety, and desaturation [174]. However, sedation interruption has been further demonstrated to reduce the complications of prolonged mechanical ventilation [176]. The SHEA/IDSA guideline recommends daily assessment of the readiness to wean and the use of weaning protocols [171]. For children, daily assessment of readiness to extubate should be carried out, but sedation interruption is not recommended because of the high risk of unplanned extubation [177].

Elevation of the Head of the Bed

Reducing the risk of aspiration and contamination with gastric secretions also helps to prevent the development of pneumonia. Positioning the head of the bed at an angle of 30 to 45 degrees reduces the risk of aspiration significantly [149,178,179]. In one randomized, controlled trial, there were 18% fewer cases of VAP among intubated patients in the group assigned to the recumbent position (45 degrees) compared with the group assigned to the supine position [179]. In another study, elevation of the head of the bed to 30 degrees was the most effective measure among a group of preventive interventions, resulting in a 52% variance in the rate of VAP [180]. Both the ATS/IDSA and SHEA/IDSA guidelines recommend maintaining the head of the bed at a 30- to 45-degree angle [28,171]. An angle of 30 to 45 degrees is also recommended for infants and children, but a lower angle (15 to 30 degrees) should be used for neonates [177].

Daily Oral Care with Chlorhexidine

Oral care interventions have been suggested by some, in part because of an association between a high level of dental plaque and a high rate of colonization with aerobic pathogens, including S. aureus, gram-negative bacilli, and P. aeruginosa[181]. Research has shown that oral decontamination with chlorhexidine leads to a significant reduction in the colonization of pathogens in the oropharynx. In most studies, the intervention has not had a significant effect on the rate of VAP or associated mortality, but more recent studies have shown a significant decrease in the rate of pneumonia [180,182,183,184,185,186]. Brushing the teeth with chlorhexidine does not seem to add benefit [183]. Regular oral care with an antiseptic solution or chlorhexidine is recommended in the ATS/IDSA and SHEA/IDSA guidelines [28,171].

Prophylaxis of Peptic Ulcer Disease

Prophylaxis of peptic ulcer disease has evolved with some conflicting views. Antacids, histamine2 receptor antagonists, and sucralfate have been traditionally given to patients receiving mechanical ventilation to prevent the formation of stress ulcers. However, reducing the amount of gastric acid can increase the risk of colonization of gram-negative bacilli in the stomach. As a result, the WHO recommends avoiding the use of these agents [187]. The CDC notes that there was insufficient evidence on the use of peptic ulcer prophylaxis and includes no recommendations in this regard in its guideline [93]. The ATS/IDSA guideline states that the risks and benefits of prophylaxis should be weighed carefully [28]. The most recent guideline, developed by SHEA/IDSA, notes that histamine2 receptor antagonists and PPIs should be avoided in patients who are not at high risk for developing a stress ulcer or stress gastritis [171]. However, peptic ulcer prophylaxis is recommended for children, as appropriate for age and health status [177].

Prophylaxis of Deep Venous Thrombosis

There is no clear relation between prophylaxis of deep vein thrombosis and VAP pneumonia, but the American College of Chest Physicians reported a decrease in the rate of VAP when such prophylaxis was implemented as part of a package of interventions and included this measure in its clinical practice guideline [188]. This recommendation also applies to children, as appropriate for age and health status [177].

Other Measures

In addition to the interventions in the ventilator bundle, other measures have been recommended to help prevent VAP. One such measure is selective decontamination of the digestive tract, which involves the use of either topical antiseptic, oral antibiotics, or a brief course of systemic antibiotics [26]. A meta-analysis of 28 studies showed that selective decontamination of the digestive or respiratory tract with use of topical antiseptic or antimicrobial agents helped reduce the frequency of VAP in the ICU [146]. The estimate of efficacy in prevention was 27% for antiseptics and 36% for antibiotics. Neither had an effect on mortality. This intervention is recommended in the SHEA/IDSA guideline [171].

Other preventive measures are targeted primarily to the care and use of ventilator equipment and practices in direct patient care. Meticulous attention to aseptic care of the equipment is necessary, and all reusable components, such as nebulizers, should be disinfected or sterilized. Tubing circuits should be replaced after 48 hours or earlier if there are signs of malfunction or contamination [93]. Changes in the design of the endotracheal tube have also been evaluated; for example, a tube with a suction port above the cuff allows for continuous aspiration of subglottic secretions. Use of this specially designed endotracheal tube has led to significantly lower rates of VAP as well as shorter durations of ventilation and shorter stays in the ICU [189,190]. Among patients who had major cardiac surgery, the greatest benefit was found for patients who received ventilation for more than 48 hours [190]. Although the cost of the tube is higher than traditional tubes, the overall cost savings in preventing VAP more than compensates [189]. In one meta-analysis, subglottic secretion drainage was significantly associated with a decreased incidence of VAP, shorter time on mechanical ventilation, and longer time to the development of pneumonia [191]. The CDC, the ATS/IDSA, and the SHEA/IDSA guidelines recommend subglottic secretion drainage with this tube when possible [28,93,171].

The use of noninvasive ventilation is another measure that has reduced the incidence of VAP [93,192,193,194]. In one study, the incidence decreased from 20% to 8% when noninvasive techniques were used routinely for critically ill patients with acute exacerbation of COPD or severe cardiogenic pulmonary edema [192]. Again, the CDC, the ATS/IDSA, and the SHEA/IDSA guidelines recommend the use of noninvasive ventilation when possible [28,93,171].

Quality Improvement Initiatives and Enhanced Infection Control Strategies

Quality improvement and infection control initiatives and strategies have led to a substantial decrease in the rates of VAP since the early 2000s [195]. The use of physician-led multidisciplinary rounds with team decision-making, checklists, and a focus on the ventilator bundle has led to significant reductions in the risk for pneumonia [196,197,198]. Strong downward trends were also found for the average length of stay in the ICU and the financial costs per patient [196].

As with HAP, strategies to decrease or eliminate modifiable risk factors for nursing home-acquired pneumonia should be implemented. A multidisciplinary panel made three recommendations for prevention of pneumonia among nursing home residents [199]:

The vaccination status of healthcare workers has been found to have a direct effect on transmission of influenza virus to patients. Outbreaks of influenza in healthcare settings have been associated with low rates of vaccination among healthcare workers, and lower rates of nosocomial influenza have been related to higher vaccination rates among healthcare workers [200,201]. Because of these findings, the ACIP recommends annual influenza vaccination for all healthcare workers, and the IDSA/ATS guideline endorses this recommendation [47]. The ACIP notes that the TIV is preferred over LAIV for workers who are in close contact with severely immunosuppressed people requiring protective isolation [112]. In addition, the Joint Commission began including vaccination programs in its accreditation standards in 2007 [123].

Despite these recommendations, only 29% to 69% of healthcare workers receive the influenza vaccination each year [202,203,204]. Healthcare workers have given many reasons for not being vaccinated, and the reasons vary among professions. Across all categories, shortage of the vaccine is the primary reason for not being vaccinated; other reasons include concern about side effects, inconvenience, and forgetfulness [204].

Efforts to increase the vaccination rate among healthcare workers are ongoing. A CDC guideline includes four level I recommendations to help increase rates of vaccination [205]:

Hand hygiene is the most important preventive measure in hospitals, and the Joint Commission mandates that hospitals and other healthcare facilities comply with the Level I recommendations in the CDC guideline for hand hygiene [206]. The CDC guideline states the specific indications for washing hands, the recommended hand hygiene techniques, and recommendations about fingernails and the use of gloves [207]. The guideline also provides recommendations for surgical hand antisepsis, selection of hand-hygiene agents, skin care, educational and motivational programs for healthcare workers, and administrative measures.

Despite the simplicity of the intervention, its substantial impact, and wide dissemination of the guideline, compliance with recommended hand hygiene has ranged from 16% to 81%, with an average of 30% to 50% [207,208,209,210,211,212]. Among the reasons given for the lack of compliance are inconvenience, understaffing, and damage to skin [207,210,213]. The development of effective alcohol-based handrub solutions addresses these concerns, and studies have demonstrated that these solutions have increased compliance [211,214,215]. The CDC guideline recommends the use of such solutions on the basis of several advantages, including [207]:

The guideline suggests that healthcare facilities promote compliance by making the handrub solution available in dispensers in convenient locations (such as the entrance to patients' room or at the bedside) and provide individual pocket-sized containers [207]. The handrub solution may be used in all clinical situations except for when hands are visibly dirty or are contaminated with blood or body fluids. In such instances, soap (either antimicrobial or nonantimicrobial) and water must be used.

However, there are many other reasons for lack of adherence to appropriate hand hygiene, including denial about risks, forgetfulness, and belief that gloves provide sufficient protection [207,210,213]. These reasons demand education for healthcare professionals to emphasize the importance of hand hygiene. Also necessary is research to determine which interventions are most likely to improve hand-hygiene practices, as no studies have demonstrated the superiority of any intervention [216]. Single interventions are unlikely to be effective.