Pneumonia is the sixth leading cause of death, and the number 1 cause of death from infectious disease, in the United States. [1, 2, 3, 4, 5, 6, 7] The image below depicts typical bacterial pneumonia.
Typical versus atypical syndromes
The classification of pneumonias as either typical or atypical arose from the observation that the presentation and natural history of some patients with pneumonia were different compared with those with pneumococcal infection.
Pathogens like Haemophilus influenzae, Staphylococcus aureus, and gram-negative enteric bacteria cause clinical syndromes similar to that due to Streptococcus pneumoniae. However, other pathogens cause an atypical pneumonia syndrome, and this was initially attributed to Mycoplasma pneumoniae.
Other pathogens, including bacteria and viruses are now known to cause similar syndromes indistinguishable from that due to M pneumonia. Therefore, the term atypical pneumonia represents diverse etiologic entities and may have limited clinical value.
Chest radiography with posteroanterior and lateral views is the preferred imaging examination for the evaluation of typical bacterial pneumonia. 
When patients present with fever, chills, or cough, pneumonia is suggested on the basis of focal or diffuse opacities.
Controversy exists with regard to the time required for an opacity to appear on chest radiographs. The vast majority of opacities appear within 12 hours. When patients are referred from the community to the radiologist, adequate time has usually lapsed for its detection. However, when nosocomial pneumonia is suspected, these patients may undergo chest radiography within a few hours, when opacities may not yet be visible on radiographs. [9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20]
In immunosuppressed patients, especially those with coexistent neutropenia, diabetes, alcoholism, or uremia, the appearance of infiltrates may also be delayed.
Other findings that suggest the presence of pneumonia include air bronchograms; the silhouette sign; parapneumonic effusions; and complications of pneumonia, such as lung abscesses, and atelectasis.
Findings that have been associated with an increased mortality, as shown in the multicenter trial by Hasley and colleagues, are bilateral pleural effusion and multilobar pneumonia.
Resolution of radiographic findings
The change in infiltrates on chest radiographs is not necessarily correlated with the activity of clinical disease. In some patients, chest infiltrates may worsen with the start or treatment, despite clinical improvement.
Pneumonia that is slow to resolve after appropriate antibiotic therapy can be a problem. Nonresolving pneumonia has been variously defined by Amberon in 1943, Henden in 1975, and Fein and colleagues in 1987 and 1993. In general, this entity is thought to be present when a patient does not improve clinically or when a radiographic infiltrate resolves slowly despite adequate and appropriate antibiotic therapy. About 10% of diagnostic bronchoscopy procedures and 15% of pulmonary consultations are performed to evaluate a nonresolving infiltrate.
The most common cause of unnecessary invasive evaluation is a failure to appreciate the length of time that infections need to clear radiologically. Studies have shown that impaired host defenses are more important determinants of delayed resolution than the infecting pathogen.
Host factors responsible for delayed resolution of pneumonia include age older than 50 years, smoking; and chronic illnesses, such as diabetes mellitus, renal failure, chronic obstructive pulmonary disease (COPD), and alcoholism.
Bacterial pneumonias usually tend to be unilobar and have cavitary lesions and effusions. Atypical pathogens can cause multilobar involvement with nodular or reticular infiltrates, lobar or segmental collapse, or perihilar adenopathy.
S pneumoniae pneumonia
S pneumoniae causes 10-50% of all cases of community-acquired pneumonia (CAP). Radiographic consolidation of the alveoli begins in the peripheral airspaces, as in the image below. The disease usually causes a lobar or segmental pattern, and a patchy bronchopneumonic pattern involving the lower lobes is seen in the elderly. A striking characteristic of S pneumoniae infection is its tendency to involve the pleura. Parapneumonic effusions are common in pneumococcal pneumonia. 
In patients with bacteremic pneumococcal patients, 50% had clear radiographs at 9 weeks, compared with 5 weeks in nonbacteremic pneumococcal pneumonia.
In patients older than 50 years with both alcoholism and COPD, 60% have an abnormal chest radiograph at 14 weeks. In patients younger than 50 years with bacteremia and no underlying illness, 40% have an abnormal chest image at 2 weeks. For the group as a whole, 37% have residual consolidation at 4 weeks, with complete resolution by 18 weeks in almost all patients.
Despite therapy during the initial phase of illness, 52% of bacteremic patients compared with 26% of nonbacteremic patients had radiographs showing deterioration. Jay and colleagues recommended that an appropriate interval for serial radiographic examination is 6 weeks, unless otherwise indicated by a patient's worsening clinical status.
H influenzae pneumonia
H influenzae pneumonia, shown in the image below, is commonly seen in COPD patients who are smokers; in the elderly; and in those with alcoholism diabetes, sickle cell anemia, or immunocompromise. This organism can be present in up to 38% of outpatients and 10% of hospitalized patients with CAP.
In most patients, radiographs demonstrate a patchy bronchopneumonic pattern, but segmental and lobar consolidation may be seen. Therefore, H influenzae pneumonia is indistinguishable from pneumococcal pneumonia. Pleural effusion is a common finding.Radiographs usually show a multilobar infiltrate and pleural effusions in 50% of cases. Resolution is usually slow.
Klebsiella pneumoniae pneumonia
The radiographic patterns seen in Klebsiella pneumonia include patchy bronchopneumonia and dense lobar consolidations. The alveoli are filled with large amounts of fluid and mucoid suppurative exudates that may cause the volume of the affected lung to increase with bulging of the interlobar fissures, has a rare feature. Although these findings are thought to be characteristic of Klebsiella pneumonia, they may be seen in other causes of pneumonia.
There is a strong tendency for abscess formation as well as pleural involvement. Cavities may develop rapidly after the onset of illness, and these may be associated with massive lung gangrene.
Pseudomonas aeruginosa pneumonia
P aeruginosa pneumonia has a characteristic predilection for the lower lobes. Patchy bronchopneumonia or extensive consolidation may be present. Involvement may be unilateral or bilateral and extensive. Extensive necrosis may be seen, with the formation of parenchymal abscesses. Massive bilateral consolidation is usually associated with a poor prognosis. Nodular infarcts may occur in the lung parenchyma.
S aureus pneumonia
This type of pneumonia may be seen as a complication of influenza, particularly during an epidemic. S aureus pneumonia usually begins in the peripheral airways rather than in the acini proper. In adults, patchy bronchopneumonia is more common and often bilateral, though lobar consolidation may be seen. Late development of abscesses is relatively common. When staphylococcal pneumonia occurs as a complication of influenza, it is usually rapidly progressive with extensive bilateral pneumonia that resembles pulmonary edema.
In children, it is usually a lobar or multilobar consolidation, rapidly progressing with the development of pneumatoceles and/or empyema. The presence of pneumatoceles in children is virtually diagnostic of staphylococcal pneumonia. Rapid progression is seen with lobar or multilobar consolidation. Pneumatoceles may rapidly develop, and empyema is frequent. 
Degree of confidence
In patients with underlying structural lung disease, the appearance of the various signs of pneumonia may not be straightforward.
Narrowing the differential diagnosis of pneumonia into typical and atypical forms on the basis of radiographic appearance alone is not reliable, as shown in a prospective study by Fang et al. 
Computed tomography (CT) scanning is increasingly used in clinical practice, but various groups have questioned its usefulness in evaluating pneumonia. Their reports have suggested that its usefulness in the diagnosis of pneumonia is limited to the following settings:
Evaluation of an indistinct, abnormal opacity depicted on a chest radiograph
Assessment of patchy, ground-glass, or linear/reticular opacities on chest radiographs
Confirmation of pleural effusion
Examination of neutropenic patients with fever of unknown origin (with the use of ultra–thin-section CT scanning)
In clinical practice, coinfection with multiple organisms is not rare, and underlying abnormalities of the lung parenchymal usually predispose patients to pneumonia. Hence, the overall clinical and radiologic picture must be considered. 
CT scans of typical bacterial pneumonia are provided below.
The literature indicates that ultrasonography can aid in the differentiation of consolidation and effusion. Consolidated lung tissue may appear as hypoechoic areas with blurred margins. The texture varies with the amount of aeration, being more heterogeneous with aeration and homogenous with dense consolidation.  Ultrasonography may also help in diagnosing empyema and abscesses. [20, 26]
Degree of confidence
The role of ultrasonography in clinical practice is limited to the identification and quantification of parapneumonic effusions. This area can then be marked for subsequent diagnostic or therapeutic thoracentesis.