Pediatric Radiology

, Volume 40, Issue 11, pp 1768–1773

Radiological findings of community-acquired methicillin-resistant and methicillin-susceptible Staphylococcus aureus pediatric pneumonia in Hawaii


    • Department of PediatricsUniversity of Hawaii, John A. Burns School of Medicine
  • Lora Bergert
    • Department of PediatricsUniversity of Hawaii, John A. Burns School of Medicine
  • Kyra Len
    • Department of PediatricsUniversity of Hawaii, John A. Burns School of Medicine
  • Marian Melish
    • Department of PediatricsUniversity of Hawaii, John A. Burns School of Medicine
  • Kevin Kon
    • Department of RadiologyKapiolani Medical Center for Women and Children
  • Robert DiMauro
    • Department of RadiologyKapiolani Medical Center for Women and Children
Original Article

DOI: 10.1007/s00247-010-1680-0

Cite this article as:
Erdem, G., Bergert, L., Len, K. et al. Pediatr Radiol (2010) 40: 1768. doi:10.1007/s00247-010-1680-0



Community-acquired Staphylococcus aureus (CA-SA) infections are common among pediatric patients in Hawaii.


We wanted to characterize the radiological features of methicillin-susceptible (CA-MSSA) and methicillin-resistant (CA-MRSA) staphylococcal pneumonia in Hawaiian children.

Materials and methods

We retrospectively reviewed medical records and imaging studies of children with SA pneumonia identified from 1996 through 2007.


Of 40 children, 26 (65%) had CA-MRSA pneumonia and 14 patients (35%) had CA-MSSA pneumonia. CA-MRSA patients were significantly younger than CA-MSSA patients (65% younger than 1 year vs. 36% older). In a majority (62%) of CA-MRSA patients, the consolidation was unilateral; in most of the CA-MSSA cases (79%), the consolidation was bilateral. Fifty percent of the patients with CA-MRSA and 21% of those with CA-MSSA had pneumatoceles (P = 0.1). CA-MRSA patients more commonly had pleural effusions (85% vs. 64% for CA-MSSA) and pleural thickening (50% vs. 36% for CA-MSSA).


This case series describes the radiologic characteristics of CA-MRSA and CA-MSSA pneumonia in children in a highly endemic area. We found that CA-MRSA pneumonias are unilateral in a majority of pediatric pneumonia cases, are more common in children 1 year or younger, and have higher rates of complications in comparison to CA-MSSA patients.


Staphylococcus aureusPneumoniaChildrenMethicillin resistance


During the last few decades, Staphylococcus aureus (SA) has emerged as a major pathogen in nosocomial and community-acquired infection in both adult and pediatric populations [18]. The clinical and bacteriologic characteristics of methicillin-resistant (MRSA) and methicillin-susceptible Staphylococcus aureus (MSSA) have been shown to be different in several studies [912]. Although staphylococcal infections are known to cause necrotizing pneumonia and secondary findings such as pneumatoceles, the radiological features of MRSA and MSSA pneumonia in children have not been well-characterized. CA-SA infections are commonly observed as a culture-proven bacterial etiology for pediatric patients in Hawaii. Among the severe and possibly life-threatening infections, SA pneumonia makes up an important portion of these admissions (MedMined from February 2003 to present, Kapiolani Medical Center). These observations have prompted us to perform a retrospective review of the medical records of children with SA pneumonia identified from 1996 through 2007 to determine the radiological characteristics of pneumonias associated with CA-MSSA and CA-MRSA infections.

Materials and methods

This study was approved by the Institutional Review Board of Hawaii Pacific Health. We performed a retrospective review of all children ages 1 month to 18 years hospitalized at Kapiolani Medical Center for Women and Children (KMCWC) between Jan. 1, 1996, and Dec. 31, 2007, with diagnosis of community-acquired staphylococcal pneumonia. KMCWC, with its 82 pediatric and neonatal beds, is the only pediatric tertiary-care facility to serve all of the civilian Hawaiian population, as well as some of the Pacific Islander populations including those from American Samoa and the Marshall Islands. Patients with multiple organisms isolated from their cultures were excluded. Children who had SA isolated from a sputum or tracheal aspirate were only included if they had clinical and radiological findings consistent with pneumonia.

Medical records and microbiology reports were reviewed to classify SA infections as CA-MRSA or CA-MSSA. The microbiology records were reviewed if the children with bilateral consolidation had SA bacteremia. Hospital-acquired SA (HA-SA) cases were excluded. HA-SA was defined as patients with SA infections who had any of the following risk factors: previous SA infection or colonization, hospitalization >48 h before isolation of SA, presence of a percutaneous device or indwelling catheter at the time of culture, or residing in a long-term care facility within the year before SA isolation. Children with SA infections who lacked documented HA-SA risk factors were classified as community-acquired (CA-SA) cases. In addition to the demographic characteristics, total length of stay, presence of influenza virus coinfection, pediatric intensive care unit admission, duration of fever, culture results, surgical and non-surgical treatment modalities, complications of infection and treatment complications were noted. The reports of chest radiographs, CT scans and any additional radiological imaging reports were reviewed. The CT scans were obtained to further characterize the imaging findings on the chest radiographs owing to clinical deterioration (increased dyspnea, oxygen requirements, persistent fever beyond 48 hours of antimicrobial treatment), and to evaluate for possible empyema and/or abscess formation. The CT scans were performed on a GE Advantage single-axial-slice scanner (GE Healthcare, Waukesak, WI, USA) (1996–2001), a GE Volume Light-Speed 16-slice scanner (December 2001–March 2007) and a GE Volume LightSpeed 64-slice scanner, Waukesak, WI (March 2007–December 2007). The chest CTs were performed using a standard protocol of 100 mA with rotation time of 0.4 s at 120 kVp and a pitch factor of 1.375. The CTDI (vol) for this was 5.17 mGy, based on the pediatric body phantom (16-cm diameter).

The presence of consolidation and its respective distributions (unilateral or bilateral) were recorded and the microbiology records were reviewed if they had SA bacteremia. The presence of pneumatoceles, nodules, cavitation within nodules or areas of consolidation, pleural thickening, pleural effusion (size, presence of loculations or intrapleural gas), and lymphadenopathy (mediastinal, hilar) was evaluated. Consolidation was defined as homogeneous increased lung attenuation that obscures the margins of vessels and airway walls. These parameters were then compared between patients with CA-MSSA and those with CA-MRSA infections. Statistical analyses were performed using SAS 9.1.3 software (SAS Institute, Cary, N.C., USA). Differences in SA characteristics were assessed using Fisher’s exact test for categorical data and the Student’s t test for continuous data.


Twenty-six (65%) children had CA-MRSA infection and 14 (35%) had CA-MSSA infection. The cultures were obtained from pleural fluid (26 children), blood (ten children), sputum (one child) and tracheal aspirate (six children). Some children had more than one test (Table 1). Twenty-three patients were younger than 1 year and 18 (78%) of these 23 had CA-MRSA infection. The information for influenza admission was available in only eight patients admitted from 2002 through 2007, and none of these patients had influenza infection.
Table 1

Demographic and culture characteristics of children with CA-MRSA and CA-MSSA pneumonia


MSSA (n = 14)

MRSA (n = 26)


Age <1 yr

5 (36%)

18 (69%)

Male, n (%)

7 (50%)

12 (46%)

Culture results

Pleural fluid

6 (43%)

20 (77%)


5a (36%)

5 (19%)

Tracheal aspirate

3 (21%)

3b (11%)


1 (1%)


a1 MSSA patient had positive pleural and blood culture; b2 MRSA patients had positive tracheal aspirate cultures with a 2nd positive culture (blood and pleural).

All 40 children had chest radiographs and 27 children had chest CT scans. Of these 27, 16 had CA-MRSA and 11 had CA-MSSA pneumonia.

There was no lobar predominance. Sixteen (62%) of the total 26 CA-MRSA patients had unilateral consolidation on admission and during their follow-up evaluations (Table 2). In contrast, the majority of the patients with CA-MSSA infection (11 of 14 total patients, 79%) had bilateral consolidation (Fig. 1). Nine of the 21 patients with bilateral consolidation had documented SA bacteremia and five of these had CA-MSSA infection. One child with unilateral pneumonia had CA-MRSA bacteremia.
Table 2

Radiological findings in children with Staphylococcus aureus pneumonia

Radiological findings

MRSA (26 patients)

MSSA (14 patients)

Fisher exact test (P-value)

Total patients (n = 40)

Unilateral consolidation

16 (≤1 year: 12  > 1 year: 4)

3 (≤1 year: 1 > 1 year: 2)



Bilateral consolidation

10 (≤1 year: 6  > 1 year: 4)

11 (≤1 year: 4 > 1 year: 7)




13/26 (50%)

3/14 (21%)



Pleural effusion

22 (85%)

9 (64%)



Pleural thickening

13 (50%)

5 (36%)



Pulmonary nodules

3 (12%)

2 (15%)



Pulmonary abscess

1 (4%)

1 (7%)




11 (42%)

3 (21%)




1 (4%)



Fig. 1

CA-MSSA pneumonia. a Frontal chest radiograph in a 9-year-old demonstrates a dense left lower lobe infiltrate as well as a small left pleural effusion. Additional hazy and rounded densities are seen bilaterally and represent additional infiltrates and multiple septic emboli. b Bilateral infiltrates in a 6-year-old. Frontal radiograph shows dense consolidation of the left lower lobe, patchy infiltrates on the right and a small left pleural effusion

Sixteen (40%) children had evidence of pneumatoceles at admission or during their follow-up radiological evaluations (Fig. 2). Eleven of these (69%) were observed in children younger than 1 year of age. The mean time from the onset of symptoms including fever, cough and respiratory distress to pneumatocele development was 12.5 days (range 7–25 days). The pneumatoceles were identified in two children by CT scans, but not by chest radiographs.
Fig. 2

CA-MRSA pneumonia in a 10-month-old. a Portable radiograph shows a right middle lobe infiltrate and a small right pleural effusion. The left lung is clear. b Follow-up chest radiograph of the same child 17 days later shows development of pneumatoceles at the right lung base

Twenty-two of the 26 CA-MRSA patients had pleural effusion, compared to 9 of the 14 CA-MSSA patients (Table 2). Thirteen of the 26 CA-MRSA patients had pleural thickening in comparison to 5 of the 14 CA-MSSA patients. None of the patients had hilar lymphadenopathy. Two patients with bacteremia and CA-MRSA died (P = 0.53). Local or systemic complications including pneumothorax, fistula formation, pneumomediastinum, rib osteomyelitis, decortication, and sepsis with multi-organ failure were more frequent among CA-MRSA infected patients (Fig. 3). Only CA-MRSA patients developed pneumomediastinum (one child), fistula (two children) and rib osteomyelitis (one child); one child with pneumomediastinum also had a fistula. Four MRSA-infected children had diagnoses of sepsis and/or endocarditis and two MSSA-infected children had sepsis/endocarditis. Two MRSA-infected children needed decortication surgery during hospitalization (P = 0.53). Overall, 18 (69%) of the CA-MRSA-infected children had local and systemic complications. Five (36%) of the CA-MSSA-infected children had complications.
Fig. 3

CA-MRSA pneumonia in a 2-month-old. a Frontal chest radiograph shows right basilar infiltrate and right-side hydropneumothorax. b Subsequent chest CT scan obtained 2 days after the chest radiograph shows diffuse pleural thickening and a hydropneumothorax on the right. There is dense consolidation of the right lower lobe (thin arrows) as well as areas of lung necrosis (thick arrows)


The present study to our knowledge is the first retrospective review that compares the radiological findings of pediatric staphylococcal pneumonia patients from a high-prevalence area. CA-MSSA and CA-MRSA are well described but uncommon causes of community-acquired pneumonia, representing 1% to 5% of reported pediatric and adult pneumonia cases [13, 14]. Recently, increasing reports of severe CA-MRSA pneumonia in young, healthy adults and children have been reported, paralleling the increasing incidence of CA-MRSA skin and soft-tissue infections [1524]. Yet, the findings of CA-SA pneumonia, MRSA and MSSA reported in the literature are limited to descriptions of necrotizing pneumonia without detailed analysis [14, 16, 18, 2024]. Early recognition of SA pneumonia is also critical in appropriate management of this infection. Radiologists and clinicians would benefit from associated radiological findings specific to CA-MRSA and CA-MSSA pneumonia.

Nguyen et al [25], in their review of nine adults with community-associated MRSA pneumonia, reported extensive bilateral consolidation and frequent cavitation [25]. In another adult study of mostly nosocomially acquired pneumonia patients, consolidation patterns were described as alveolar, nodular and mixed [26]. In that study no difference among these patterns or presence of bilateral involvement was found between MRSA- and MSSA-infected patients. A Medline (PubMed) search of English literature since 1964 with the search words of staphylococcus, pneumonia and radiology did not reveal any other reports of case series or any radiological comparisons between MSSA and MRSA pneumonias.

We believe this is a unique series describing the radiological characteristics of MRSA and MSSA pneumonia in children in a highly endemic area [27, 28]. Our study was somewhat limited by the small number of patients with community-acquired pneumonia. However, there were significant differences between the radiological features of MRSA- and MSSA-associated pneumonias and their complications. We found that MRSA infections are mostly unilateral in distribution and more common in very young children. In contrast, MSSA infections are mostly bilateral in distribution and are common in older age groups. The difference is statistically significant. We also found the rate of pleural effusions and pneumatoceles was higher among MRSA-infected children. We have observed the radiological appearance of pneumatoceles as early as 7 days after the onset of symptoms [29].

MRSA pneumonia has been reported to have a more complicated course [3032]. A retrospective chart review of staphylococcal pneumonia patients, including the ones in this report, showed that MRSA-infected patients suffer from more frequent pleural effusions, pneumatocele formation, complications and deaths (paper in progress). The bacteriologic explanations for invasiveness and possible clinical differences between MRSA- and MSSA-associated infections are not well known. The invasiveness has been associated with certain clonal types of staphylococci such as USA300 and staphylococcal bacterial toxins including Panton-Valentine leukocidin [33]. There have also been contrasting reports showing no significant association [34]. In this study we did not have the bacterial isolates to analyze the clonal types and virulence factors. In a more invasive infection, MSSA or MRSA could cause further tissue damage and might result in further pneumatocele formation. Unfortunately, the information that might be gathered from bacterial analyses might still not explain why our MRSA-infected patients had more unilateral consolidation in comparison to MSSA-infected patients.


Early consideration of staphylococcal pneumonia is especially important in a time when the rates of community-acquired MRSA infections and their complications have been steadily increasing. Although there was no single radiological feature to differentiate MRSA and MSSA infections in our study, we have observed significant differences. Radiological findings alone should in no way be used in place of the clinical and bacteriologic information. Considering that many patients have chest radiographs before culture and susceptibility testing, the reported radiological findings remain helpful for diagnosis, treatment and follow-up for children with staphylococcal pneumonia.

Copyright information

© Springer-Verlag 2010