European Journal of Radiology 51 (2004) 102113

Radiology of bacterial pneumonia

Jos Vilar , Maria Luisa Domingo, Cristina Soto, Jonathan Cogollos
Radiology Department, Hospital Universitario Doctor Peset, Valencia, Spain Received 23 February 2004; received in revised form 26 February 2004; accepted 1 March 2004

Abstract Bacterial pneumonia is commonly encountered in clinical practice. Radiology plays a prominent role in the evaluation of pneumonia. Chest radiography is the most commonly used imaging tool in pneumonias due to its availability and excellent cost benet ratio. CT should be used in unresolved cases or when complications of pneumonia are suspected. The main applications of radiology in pneumonia are oriented to detection, characterisation and follow-up, especially regarding complications. The classical classication of pneumonias into lobar and bronchial pneumonia has been abandoned for a more clinical classication. Thus, bacterial pneumonias are typied into three main groups: Community acquired pneumonia (CAD), Aspiration pneumonia and Nosocomial pneumonia (NP).The usual pattern of CAD is that of the previously called lobar pneumonia; an air-space consolidation limited to one lobe or segment. Nevertheless, the radiographic patterns of CAD may be variable and are often related to the causative agent. Aspiration pneumonia generally involves the lower lobes with bilateral multicentric opacities. Nosocomial Pneumonia (NP) occurs in hospitalised patients. The importance of NP is related to its high mortality and, thus, the need to obtain a prompt diagnosis. The role of imaging in NP is limited but decisive. The most valuable information is when the chest radiographs are negative and rule out pneumonia. The radiographic patterns of NP are very variable, most commonly showing diffuse multifocal involvement and pleural effusion. Imaging plays also an important role in the detection and evaluation of complications of bacterial pneumonias. In many of these cases, especially in hospitalised patients, chest CT must be obtained in order to better depict these associate ndings. 2004 Elsevier Ireland Ltd. All rights reserved.
Keywords: Pneumonia; Bacterial pneumonia; Pulmonary CT; Nosocomial pneumonia

1. Introduction Bacterial pneumonias account for a large percentage of all pneumonias. They have been classied into three main groups: lobar pneumonia, bronchopneumonia and acute interstitial pneumonia [1]. Lobar pneumonias are characterised by conuent areas of focal airspace disease, usually limited to one lobe or segment. Bronchopneumonia has a multifocal distribution with nodules that tend to join producing air-space consolidations affecting one or more lobes. Acute interstitial pneumonias are produced by involvement of the bronchial and bronchiolar wall, and of the pulmonary interstitium, and are most commonly caused by viral organisms and Mycoplasma pneumoniae. This classic morphologic classication is of limited usefulness because the radiographic pattern often cannot be used to predict the causative organism. The appearance of new infective organisms, the increasing age of the population and the wide use of antibiotics have changed the pat

terns of this disease [2]. This is why most authors prefer to classify pneumonias from the perspective of the mechanism of origin. Thus, we will refer to three main groups of pneumonias: community acquired pneumonia (CAP), nosocomial pneumonia (NP) and aspiration pneumonia. Streptococcus pneumoniae is the most common cause of CAP while Gram-negative bacteria and Staphylococcus aureus are more often responsible for hospital acquired pneumonia [2]. Aspiration pneumonias are usually produced by micro-organisms that colonize the oropharynx which include Gram-positive cocci, Gram-negative rods, and rarely, anaerobic bacteria. This article will review the most common and some unusual radiographic presentations of bacterial pneumonia in inmunocompetent patients.

2. Imaging pneumonia In patients with suspected pneumonia, imaging plays a major role in the detection, characterisation and follow-up of the disease.

Corresponding author. E-mail address: vilar jlu@gva.es (J. Vilar).

0720-048X/$ see front matter 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejrad.2004.03.010

J. Vilar et al. / European Journal of Radiology 51 (2004) 102113

103

2.1. Detection The basic and most diffused imaging tool to diagnose pneumonia remains the chest radiograph. Indeed pulmonary infections are the most common reason for obtaining an emergency chest lm. Pneumonia may present with a wide spectrum of symptoms and often the initial clinical manifestations are clear. Although the chest radiograph is

often regarded as the reference standard for the diagnosis of community-acquired pneumonia, its reliability is limited by signicant interobserver variability in radiographic interpretation [3]. Other techniques like computed tomography (CT) can be useful, showing some inltrates not visualised in the chest radiographs (Fig. 1) and can assure the existence of cavitation or other complications, [4] but the use of CT is only

Fig. 1. Additional value of CT: CAP (a) chest radiograph: there is a paratracheal opacity in the right upper lobe. (b) CT of the same patient shows clearly the opacity due an air-space consolidation.

104

J. Vilar et al. / European Journal of Radiology 51 (2004) 102113

recommended in cases uncertain to the chest lm, complications of pneumonia or suspicion of an underlying additional lesion such as bronchogenic carcinoma. Magnetic resonance imaging (MRI) can demonstrate pulmonary consolidations. It can be used as an alternative to CT in patients who should not be exposed to ionising radiation. 2.2. Characterisation Is imaging reliable for distinguishing the infective organism? Tew et al. [5] reviewed 31 patients with bacterial and non-bacterial pneumonias. The diagnostic accuracy was 67% for bacterial pneumonia and 65% for non-bacterial pneumonia. The authors concluded that radiology alone was unable to distinguish bacterial from non-bacterial pneumonias. In a review of 114 cases of pneumonia, Reittner et al. concluded that CT is also unable to differentiate the aetiology of various types of pneumonia except Pneumocystis carinii [6]. The characterisation of some NP may be quite difcult, especially in patients with assisted ventilation when other pulmonary conditions may coincide [7]. Despite these limitations, imaging may be of great help in detecting the associated ndings. A study by Albaum et al. [3] showed that the chest radiograph reliability for detecting pleural uid and multiple inltrates was good. This is important since both ndings are related to a worse prognosis. 2.3. Follow-up Most pnemonias will resolve in 1 or 2 weeks. Slow resolution can occur when there are certain associated conditions such as chronic obstructive pulmonary disease, alcoholism, diabetes and immune-deciency. Otherwise, if the pneumonia does not resolve, an underlying pathology should be suspected, especially bronchogenic carcinoma. In these cases, as mentioned previously, CT is recommended [8,9].

3. Community acquired pneumonia (CAP) The aetiology of CAP varies widely according to the different reviews published. It is highly inuenced by the geographic area, the population studied and the diagnostic methods used [10]. The most common bacterial agents responsible for CAP are S. pneumoniae, M. pneumoniae, Chlamydia pneumoniae and Legionella pneumophila. S. aureus may complicate a viral pneumonia. CAP may be caused by Gram-negative organisms in elderly patients, alcoholics, patients with cardiopulmonary disease and due to the widespread use of broad-spectrum antibiotics [1]. The incidence of these organisms varies according to the different authors. Thus, in a study by Lim et al. [11], the most common agent producing CAP was S. pneumoniae (48%) followed by virus (19%), C. pneumoniae (13%), Haemophilus inuenzae (20%) and M. pneumoniae (3%), while another publication [2] reported S. pneumoniae

Fig. 2. Community acquired pneumonia (Streptococcus pneumoniae) (a) and (b): PA and lateral chest lms show consolidation in the lateral segment of the middle lobe, abutting the major and minor ssures.

J. Vilar et al. / European Journal of Radiology 51 (2004) 102113

105

Fig. 3. PA chest radiograph shows an alveolar consolidation involving the right and left lower lobes in a patient infected by Streptococcus pneumoniae.

(920%), M. pneumoniae (1337%) and C. pneumoniae (17%) as most common agents. The usual imaging nding in CAP coincides with the classic presentation of lobar pneumonia: an airspace consolidation in one segment or lobe, limited by the pleural surfaces (Fig. 2). CT may additionally show ground glass attenuation, centrilobular nodules, bronchial wall thickening and centrilobular branching structures [4] (Fig. 1b).

3.1. Pneumococcal pneumonia S. pneumoniae is the most frequent micro-organism causing CAP [2,11]. The usual presentation is a lobar pneumonia involving one segment or lobe. Nowadays, the use of antibiotics has changed the appearance of Pneumoccoccal pneumonia, and it may appear as patchy conuent areas that may be multilobar or bilateral (Fig. 3). Kantor [12] found

Fig. 4. Mycoplasma pneumonia: chest radiograph. There is a diffuse peripheral and bilateral interstitial involvement.

106

J. Vilar et al. / European Journal of Radiology 51 (2004) 102113

Fig. 5. (a) Legionella pneumonia: chest radiograph of a patient with fever, dyspnea and myalgias. There is a smooth bilateral perihilar consolidation. (b) Chest radiograph obtained 48 h later, notice the rapid extension of the consolidation. (c) and (d) On CT, the consolidations are multiple and bilateral.

J. Vilar et al. / European Journal of Radiology 51 (2004) 102113

107

Fig. 5. (Continued ).

108

J. Vilar et al. / European Journal of Radiology 51 (2004) 102113

that the patterns of lobar pneumonia and bronchopneumonia were equally frequent in Pneumococcal pneumonia. Another common nding in Pneumococcal pneumonia is the presence of small pleural effusions that are usually reactive. 3.2. Mycoplasma pneumonia The incidence of Mycoplasma infection is variable according to different series and may be inuenced by epidemics. Every 48 years, the incidence may reach up to 50%. This is a pneumonia of children, adolescents and adults below 40 years of age [13]. Mycoplasma pneumonia has variable radiographic appearances. In 1975, Putnan et al. [14] identied two main clinical and radiographic groups: one group had unilateral or bilateral air-space disease with a lobar or segmental distribution, while the other with a longer duration of symptoms, had a diffuse bilateral reticulo-nodular pattern (Fig. 4). A review of 31 cases of M. pneumoniae in outpatients revealed no predominant radiographic pattern (interstitial or alveolar) with more frequent involvement of the lung bases [15]. 3.3. Chlamydia pneumonia The radiographic appearance of C. pneumoniae is similar to that of M. pneumoniae, most commonly as a localised area of consolidation which may be patchy or homogeneous. Chlamydia and Mycoplasma often coexist [1]. 3.4. Legionella pneumonia Legionnella pneumophila is responsible for Legionnella pneumonia or Legionnaires disease. These infections are acquired by breathing droplets of contaminated water. The disease may be sporadic or may occur in outbreaks, most frequently in places where the population is exposed to air conditioning towers, water distribution systems and humidiers colonised by the germ [16]. The clinical features of Legionella pneumonia are typical, consisting in diarrhoea, headache, myalgias, dyspnea and cough. The radiographic ndings are often those of segmental peripheral consolidations that spread rapidly producing opacication of one or more lobes (Fig. 5). They become bilateral in half of the cases [17]. 3.5. Unusual patterns of CAP 3.5.1. Round pneumonia (Fig. 6) It was described in children but occasionally it may happen in adults. In the presence of a pulmonary nodule, round pneumonia should be suspected especially if no previous lms are available, a rapid growth is observed or there are signs of infection [18]. A variant of this could be the cases described in screening for lung cancer where some small pulmonary nodules detected will disappear after the antibiotic treatment [19]. Aspiration is the inhalation of orofaringeal or gastric contents into the larynx and lower respiratory tract. If the inhalation is of regurgitated sterile gastric contents, aspiration pneumonitis is caused; and if it is of colonised oropharingeal material, aspiration pneumonia occurs [20]. Factors that predispose to aspiration pneumonitis are those that produce disturbance of consciousness such as drug abuse, seizures, massive cerebrovascular accident, or the use of anaesthesia. Aspiration pneumonia is conditioned by neurologic disphagia, anatomic abnormalities of the upper aerodigestive tract, gastroesophageal reux in elderly persons, or poor oral care. The radiographic appearance of aspiration pneumonia and pneumonitis is variable [21] but the most common pattern is that of bilateral and multicentric opacities, particularly in the right lung, with a perihilar and basal distribution (Fig. 8).

Fig. 6. Round pneumonia: a consolidation is seen in the right lower lobe lung of this adult patient. Streptococcus pneumoniae was obtained in the sputum cultures.

3.5.2. Bilateral or multilobar pneumonia CAP can be diffuse and bilateral in patients with underlying chronic obstructive pulmonary disease due to the distortion and destruction of the pulmonary parenchyma (Fig. 7). Some of these cases will present as a linear pattern that could be confused with other aetiologies.

4. Aspiration pneumonia

5. Nosocomial pneumonias Nosocomial pneumonia or hospital acquired pneumonia is dened as a pneumonia occurring 48 h after hospital admission, excluding any infection that is incubating at the time of hospital admission, and also a pneumonia which occurs within 48 h after discharge from the hospital [22].

J. Vilar et al. / European Journal of Radiology 51 (2004) 102113

109

Fig. 7. (a) Chest radiograph of a patient with bullous emphysema. (b) The same patient with pneumonia in the left upper lobe. An airuid level (arrows) within the bullae mimics cavitation. (c) CT of this area showing the uid lled bulla.

Fig. 8. Aspiration pneumonia: chest radiograph of a patient in a comatose condition due to drug abuse. Bilateral lower lobe consolidations.

According to the literature, the incidence of NP is variable, probably because the groups of patients studied differ and the diagnostic criteria vary. These variations depend greatly on the type of hospitalisation and wards (surgical or medical). Risk factors involved in NP are the previous condition of the patient, age, severity of the underlying disease, the length of hospitalisation and the instrumentation used in invasive techniques. The most common micro-organisms responsible for NP are aerobic Gram-negative bacilli (Enterobacteriae, E. coli, Pseudomona aeruginosa), and some Gram-positive cocci such as S. aureus and S. pneumoniae. Anaerobic organisms are less common. Quite often, multiple different germs are found [23]. In patients hospitalised in Intensive Care Units, these pneumonias are more frequent, and the mortality is very high (1050%). Mechanical ventilation constitutes a great risk factor for NP since it can facilitate the growth and

110

J. Vilar et al. / European Journal of Radiology 51 (2004) 102113

Fig. 9. Ventilator assisted pneumonia: chest radiograph of a patient obtained after 5 days of mechanical ventilation. There is a right perihilar consolidation. Acinetobacter was obtained from bronchoaspirate cultures.

dissemination of germs and the cough mechanism is reduced. This has been denominated as ventilator associated pneumonia (VAP). Nevertheless, NP in the Intensive Care Units may also occur in non-ventilated patients. Thus NP has been classied in two groups: ventilator associated pneumonia and pneumonia in non-ventilated patients [24]. The incidence and mortality of the former is much higher

than that of NP in non-ventilated patients, and they also differ in their treatment. Micro-organisms responsible for VAP vary according to the duration of mechanical ventilation: VAP occurring in the rst 5 days of ventilation is usually due to S. pneumoniae, H. inuenzae or Moxarella catarrhalis and uncommonly by anaerobes, while VAP occurring after 5 days (Fig. 9) of ventilation is most commonly

Fig. 10. Nosocomial pneumonia: chest radiograph shows patchy and peripheral areas of consolidation in a hospitalised non-ventilated patient under a long-term treatment with steroids. The responsible organism was Pseudomona aeruginosa.

J. Vilar et al. / European Journal of Radiology 51 (2004) 102113

111

produced by Pseudomonas aeruginosa, Acitenobacter or Enterobacter spp., or methicillin-resistant S. aureus [25]. The radiographic pattern of NP may be quite variable These pneumonias are most commonly bilateral with diffuse or multiple foci of consolidation not limited to one lobe [7]. They may frequently associate pleural effusion (Fig. 10). The role of portable chest lms in cases of suspected NP is limited, since the presence of focal alveolar consolidations is quite frequent in these patients, and often caused by atelectasis, pulmonary infarction, oedema or acute respiratory distress syndrome (ARDS). The radiographic signs of NP are non-specic. A study by Wunderink et al. found that the only reliable sign of pneumonia was the presence of air bronchograms, except in patients with ARDS [26]. Atelectasis may solve rapidly, especially after vigorous physiotherapy. In patients with ARDS, the diagnosis of pneumonia becomes very difcult [27,28]. Generally, ARDS is bilateral, symmetric and more evident in dependent areas [29].The presence of focal areas of consolidation favours the diagnosis of pneumonia but asymmetry may also occur in ARDS [29]. Additionally, the agreement between

readers in this pathology is very low, and other factors such as the technique used to obtain the chest radiograph and the ventilator settings may inuence the results [30]. In summary, the role of radiology in NP is limited but decisive. Delay in treating pneumonia may be fatal and treating with antibiotics other entities (pulmonary infarction, oedema) may also have negative results. In hospitalized patients, the chest radiographs are most helpful when they are normal and rule out pneumonia [7]. CT may be of great help in some cases when the chest lms are inconclusive especially in patients with ARDS.

6. Complications All pneumonias, CAP and nosocomial may complicate. Complications are more common in inmunodepressed patients and in nosocomial pneumonias.

Fig. 11. Hospital acquired pneumonia: pulmonary gangrene produced by Klebsiella pneumoniae in a hospitalised patient. Notice sloughed lung tissue due to extensive necrosis in a large cavity with an airuid level.

Fig. 12. (a) Chest radiograph of a 12 months old child, with a consolidation in left lower lobe. (b) Chest radiograph obtained 4 weeks later. A cystic space has developed in the area of previous pneumonia, corresponding to a pneumatocele (arrows).

112

J. Vilar et al. / European Journal of Radiology 51 (2004) 102113

Cavitation suggests bacterial disease rather than viral or Mycoplasma infection. S. aureus, Gram-negative, anaerobic bacteria are the most common agents. Pulmonary gangrene is a rare but interesting form of cavitation that produces sloughed lung within a large cavity secondary to thrombosis of the pulmonary vessels [17]. S. pneumoniae and Klebsiella are the most common agents responsible for cavitation in inmunocompetent patients and Aspergillus in the inmunocompromised host (Fig. 11).

Pneumatocele [1] is an air cystic space that may develop as a complication of acute staphylococcal infection in children (Fig. 12). Care needs to be taken to avoid misdiagnosing cavitation and pneumatocele formation when the focal lucencies within the consolidation are due to underlying emphysema (Fig. 7). 6.1. Pleural effusion and empyema Parapneumonic effusions complicate the course of 2060% of patients hospitalised with bacterial pneumonia. Pleural effusion in CAP is less frequent and usually reactive. Most of these effusions follow an uncomplicated course and resolve with antibiotic therapy of the underlying pneumonia. In 510% cases, they become complicated and progress to empyema [31]. 6.2. Lobar enlargement This sign was well described by Felson et al. in 1949 and initially attributed to Klebsiella pneumonia (Friedlanders pneumonia) [32]. Swellling of a lobe occurs when there is an extensive exudative process. Other infectious processes such as tuberculosis and pneumococci can also demonstrate lobar enlargement (Fig. 13).

7. Conclusions Pneumonias can be classied in three main groups: community acquired pneumonia, nosocomial pneumonia and aspiration pneumonia. The role of the radiologist is to be decisive in their diagnosis and follow-up. The chest radiograph remains a basic tool for this purpose. CT is used as a complement to plain lms and especially in the evaluation of complications or unfavourable resolution of a pulmonary inltrate. The role of radiology in the intensive care unit patient is more limited since there is a great overlap of pathologies that can have similar radiographic signs. Close follow-up of these patients and adequate clinical correlation is mandatory. CT in these cases can add signicant information when portable lms are inconclusive.

References
[1] Bhalla M, McLoud TC. Pulmonary infections in the normal host. In: McLoud TC, editor. Thoracic radiology, the requisites. Mosby, USA; 1998. [2] American Thoracic Society. Guidelines for the management of adults with community-acquired pneumonia. Am J Respir Crit Care Med 2001;163(7):173054. [3] Albaum MN, Hill LC, Murphy M. Interobserver reliability of chest radiograph in community-acquired pneumonia. Chest 1996;110:343. [4] Tanaka N, Matsumoto T, Kuramitsu T, et al. High resolution CT ndings in community-acquired pneumonia. J Comput Assist Tomogr 1996;20:6008.

Fig. 13. Loefers pneumonia: (a) the chest radiograph shows an opacity in the left upper lobe. (b) Lateral chest radiograph showing posterior displacement of the major ssure due to abundant exudate by Klebsiella pneumoniae.

J. Vilar et al. / European Journal of Radiology 51 (2004) 102113 [5] Tew J, Calenoff L, Berlin B. Bacterial or nonbacterial pneumonia: accuracy of radiographic diagnosis. Radiology 1977;124:60712. [6] Reittner P, Ward S, Heyneman L, Mulle NL. Pneumonia: highresolution CT ndings in 114 patients. ECR 2001Presentation B-0937. [7] Chastre J, Fagon JY. Ventilator-associated pneumonia. Am J Respir Crit Care Med 2002;165(7):867903. [8] Johnson JL. Slowly resolving and nonresolving pneumonia: questions to ask when response is delayed. Postgrad Med 2000;108(6):115 22. [9] Burke M, Fraser R. Obstructive pneumonitis: a pathologic and pathogenic reapraisal. Radiology 1988;166:699. [10] Torres A, Soler N. Revisin de las diferentes normativas sobre el tratamiento de la neumon a adquirida en la comunidad. Med Clin Monogr 2001;2:806. [11] Lim WS, Macfarlane JT, Boswell TCJ, et al. Study of community acquired pneumonia aetiology in adults admitted to hospital: implications for management guidelines. Thorax 2001;56:296301. [12] Kantror HG. The many radiologic faces of pneumococcal pneumonia. AJR 1981;137:1213. [13] Ghosh K, Clements GB. Surveillance of Mycoplasma pneumoniae infections in Scotland 19861991. J Infect 1992;25(2):2217. [14] Putnam CE, Curtis AM, Simeone JF, et al. Mycoplasma pneumoniae: clinical and roentgenographic patterns. Radiology 1975;124:417. [15] Snchez J, Maldonado MJ, Rubio C, et al. Pneumonia due to mycoplasma pneumonia in a primary health care district. An Esp Pediatr 2001;55(2):10812. [16] Roig J, Sabria M, Pedro-Botet ML. Legionella spp.: community acquired and nosocomial infections. Curr Opin Infect Dis 2003;16(2):14551. [17] Armstrong P, Dee P. Infections of the lungs and pleura. In: Armstrong P, Wilson A, Dee P, Hansell D, editors. Imaging of diseases of the chest. 2nd ed. St. Louis; 1995. p. 145228. [18] Wagner AL, Szabunio M, Hazlett KS, Wagner SG. Radiologic manifestations of round pneumonia in adults. AJR 1998;170:7236.

113

[19] Henschke C, Naidich D, Yakelevich D, et al. Early lung cancer action project. Initial ndings on repeat screening. Cancer 2001;92(1):153 9. [20] Marik PE. Aspiration pneumonitis and aspiration pneumonia. N Engl J Med 2001;344(9):66571. [21] Katz DS, Leung AN. Radiology of pneumonia. In: Niederman MS, editor. Clin Chest Med. 1999;20(3):54962. [22] Hffken G, Niederman MS. Nosocomial pneumonia. Chest 2002;122(6):218396. [23] Marav -Poma E, Manrique A. Infecciones respiratorias nosocomiales. In: Ginestal RJ, et al, editors. Cuidados intensivos. Madrid: ELA; 1991. p. 144559. [24] Ewing S, Bauer T, Torres A. The pulmonary physician in critical care. Thorax 2002;57:36671. [25] Grossman RF, Fein A. Evidence-based assessment of diagnostic tests for ventilation-associated pneumonia. Chest 2000;117:17781. [26] Wunderink RG, Goldenberg LS, Zeiss J, Day CM, Ciemins J, Lacher DA. The radiologic diagnosis of autopsy-proven ventilator associated pneumonia. Chest 1992;101:45863. [27] Winer-Muram HT, Steiner RM, Gurney JW, et al. Ventilatorassociated pneumonia in patients with adult respiratory distress syndrome: CT evaluation. Radiology 1998;208:1939. [28] Bauer TT, Torres A. Acute respiratory dystress syndrome and nosocomial pneumonia. Thorax 1999;54:103640. [29] Desai SR, Wells AU, Suntharalingam G, Rubens MB, Evans TW, Hansell DM. Acute respiratory distress syndrome caused by pulmonary and extrapulmonary injury: a comparative CT study. Radiology 2001;218:68993. [30] Wunderink RG. Radiologic diagnosis of ventilator-associated pneumonia. Chest 2000;117:188s90s. [31] Taryle DA, Potts DE, Sahn SA. The incidence and clinical correlates of parapneumonic effusions in pneumococcal pneumonia. Chest 1978;74:170. [32] Felson B, Rosemberg LS, Hamburger M. Roentgen ndings in acute Friedlanders pneumonia. Radiology 1949;53:559.