Skip to main content

Table 2 Retrospective studies in children with necrotizing pneumonia secondary to Streptococcus pneumoniae infection

From: Necrotizing pneumonia: an emerging problem in children?

Study Subjects S. pneumoniae serotypes Imaging results Management Course Comments
Hsieh et al. [39]
Taiwan. Single tertiary center.
5 yr. review May 1998–July 2003.
No. = 15.
2 (13%) males.
Median age 49 mths (Range 9–85 mths).
10 isolates available for serotyping.
Serotype 14 (50%)
3 (30%)
6A (10%)
18C (10%).
11 isolates available for penicillin MICs.
7 (64%) had non-penicillin (MIC ≥1 mg/L) susceptibility.
CXR + CT scans performed.
Lungs involved - right 10 (67%) left 5 (33%) multi-lobar 5 (33%).
PPE 14 (93%)
PTX 1 (7%)
Procedures for PPE 14 (93%).
Chest tube drainage only 6 (43%).
VATS 8 (57%).
Median febrile days before presentation = 4 days (Range 2–11 days).
Median febrile days in hospital = 9 days (Range 4–30 days).
Median hospital LOS = 18 days (Range 5–40 days).
Mean (SD) antibiotic days for 14 survivors = 24 (8) days (Range 12–38 days).
1 died from HUS and pulmonary gangrene.
Diagnosis of NP made on a median of 6 days (range 1–12) after admission.
13 (87%) received cefotaxime or ceftriaxone 10 (67%) received vancomycin 7 (47%) received both vancomycin and a cephalosporin together as directed therapy.
Fretzayas et al. [40]
Greece.
Single tertiary center.
6 yr. review January 1999–December 2004.
No. = 10.
5 (50%) males.
Mean age 3.1 yrs.
Representing 1.3% of all cases of lobar pneumonia in children aged <14 yrs. and 20% of bacteraemic pneumonia cases.
All had +ve blood cultures, 5/7 pleural fluid cultures were also +ve.
Serotyping NR.
All had CXR evidence of NP, which was then confirmed by chest CT scans.
Epyema present in 7 patients.
Chest US aided diagnosis in all 8 cases where it was used.
Empyema drained by thoracentesis or by indwelling continuous intercostal tube drainage.
None underwent thoracotomy.
Toxic appearance, persistent fever and abnormal chest findings were seen for a mean 23 days.
Hospital LOS 15–35 days.
4 patients examined
1 yr. later had normal spirometry.
NP should be considered in those with continuation of fever, persistently raised or increasing blood inflammatory indices and abnormal auscultatory findings for >5-days, despite antibiotics and especially if bacteremia and empyema are present.
Bender et al. [29]
United States.
Single tertiary center.
9 yr. review January 1997–March 2006.
No. = 33. 18 (55%) males.
Mean age 40 mths. Co-morbidities 2 (6%).
Cases increased over time: 5/39 (13%) confirmed PNP cases in 1997–2000 vs 28/85 (33%) in 2001–2006 (OR 3.34, 95%CI 1.11–12).
Blood and pleural fluid isolates.
28/33 (85%) were non-PCV7 vaccine serotypes - 2/5 (40%) 1997–2000- 27/28 (96%) 2001–06.
Commonest serotypes - serotype 3 11 (33%)
19 4 (12%)
19A 4 (12%)
1 3 (9%).
All penicillin susceptible.
CXR or CT scans performed.
PPE 32 (97%).
Procedures for PPE 32 (93%).
Chest tube drainage only 19 (59%).
Chest tube + surgery 11 (34%).
Nursed in intensive care 18 (55%).
Mean hospital LOS = 14 days. 1 (3%) death
May have included a small number of cases with lung abscess.
Serotype 3 was most often associated with NP with 11/14 (79%) children with culture-confirmed pneumonia developing cavities.
Compared with other serotypes, serotype 3 was more likely to be associated with PNP (OR = 14.7, 95% CI 3.4–86).
Hsieh et al. [32]
Taiwan.
Single tertiary center.
9 yr. review January 2001–March 2010.
No. = 50.
18 had BPF
- 4 (22%) males - mean (SD) age 3.5 (2.7) yrs. co-morbidity 1 (6%)
PCV7 available in Taiwan since 2005, but coverage <16% by 2009.
None of the PNP cases had received PCV7.
Proportions of cases increased between 2004 and 2009.
Blood and pleural fluid isolates (n = 50).
Commonest serotypes
- serotype 14 19 (38%)
19A 10 (20%)
3 7 (14%)
6B 8 (16%).
Commonest STs - ST46 7/19 (37%) serotype 14 isolates CC320 (ST320, ST3164) 10/10 (100%) serotype 19A isolates.
6/18 (33%) with BPF had non-penicillin susceptibility and 3/18 (17%) had ceftriaxone. non-susceptibility
CXR or CT scan.
Of 18 children with BPF, all had PPE.
Bilateral lung involvement 10 (56%).
Procedures for BPF 18 children subset.
9 developed BPF after removal of chest drains.
Chest tube + surgery 15 (83%), pulmonary resection in 12.
None received fibrinolytics.
18 children with BPF.
Median febrile days before presentation = 7 days (Range 4–14 days).
Median day of BPF diagnosis = 10 days (Range 1–21 days). Nursed in intensive care 9 (50%).
Median febrile days in hospital = 13 days (Range 1–34) days.
Median hospital LOS = 32 days (Range 11–62 days).
None died.
Study focused upon 18 children with BPF, a more critically ill subset.
6 (33%) had HUS.
Multivariate analysis found acute respiratory failure (OR = 8.9, 95%CI 2.6–31) and serotype 19A (OR = 5.0, 95%CI 1.2–22) were independent risk factors for BPF.
11/12 resected lung segments had coagulative necrosis with pulmonary infarction.
Janapatla et al. [25]
Taiwan.
Single tertiary center.
3 yr. review January 2006–December 2009 Two subsets:
a. PNP
b. PNP + HUS.
No = 12 PNP.
5 (42%) males.
Mean (SD) age = 4.6 (2.4) yrs. (Range 1–8 yrs).
PCV7 available in Taiwan since 2005, but coverage <16% by 2009.
No. = 17 with PNP + HUS (+ 1 other had empyema). 4 (23%) males.
Mean (SD) age = 5.3 (2.8) yrs. (Range 2–10 yrs).
Blood and pleural fluid isolates (n = 12).
Commonest serotypes - serotype 14 9 (75%) 3 1 (8%) 19A 2 (17%)
Commonest STs - ST876 and ST46 in 7/9 (78%) serotype 14 isolates.
S. pneumoniae isolated from blood (n = 16) and pleural fluid (n = 2).
Commonest serotypes - serotype 14 9 (50%) 3 5 (28%) 19F 2 (11%).
Commonest STs - ST46 in 5/9 (56%) serotype 14 isolates ST 180 in 5/5 (100%) serotype 3 isolates.
All cases of NP confirmed by CT scan.
Results NR.
CXR or CT scan
Results NR.
Chest tube + surgery, either VATS or lobectomy, for 10 (83%) children.
 Chest tube alone 2 (11%) children, including the 1 child with empyema alone Chest tube + surgery, either VATS or lobectomy, for 16 (89%) children
Mean (SD) hospital LOS = 26.2 (9.0) days (Range 16–39 days).
Deaths, NR.
 Mean (SD) hospital LOS = 31.4 (9.0) days (Range 11–65 days). Deaths, NR.
The main focus of this study was on PNP with HUS rather than PNP alone. The 12 children with PNP and 42 with IPD without NP served as controls.
 5/18 cases of PNP + HUS had serotype 3 isolates compared with 4/54 IPD + PNP only cases (RR = 3.75, 95%CI 1.1, 12.7).
16/18 HUS isolates carried the nanC gene compared with 22/54 control isolates (RR = 2.18, 95%CI 1.52–3.13).
As 16/17 children with HUS had PNP, NanC could be a virulence factor for NP too.
Hsieh et al. [34]
Taiwan.
Six tertiary centers. 2 yr. review March 2010–April 2012.
No. = 57 with PNP.
26 (46%) males.
Mean (SD) age:
a. Mild necrosis (n = 13) 45.4 (16) mths
b. Cavitation (n = 27) 49.6 (15.7) mths c. BPF (n = 17) 39.4 (16.1) mths.
No co-morbidities.
Blood and pleural fluid positive testing by culture or PCR.
Commonest serotypes in pleural fluid:
Serotype - 19A (69%)- 3 (12.5%).
12 (21%) had respiratory viruses detected by PCR.
CXR or CT scan.
Mild necrosis (non-enhanced areas on contrast CT scans) (n = 13; 23%).
Cavitation (including pneumatoceles and abscess) (n = 27; 47%).
BPF (n = 17; 30%).
NR. Mean (SD) duration of fever and hospital LOS-
Mild necrosis: 11.4 (6) and 15.2 (4.5) days;
Cavitation: 13.9 (6.5) and 18.3 (6.4) days;
BPF: 17.8 (7.6) and 32.8 (16.8) days.
No deaths.
Pleural fluid pneumococcal load was significantly higher for serotypes 19A and 3 than other serotypes.
Severity of necrosis was associated with pleural fluid pneumococcal load and IL-8 levels.
  1. BPF bronchopleural fistula, CC clonal complex CI confidence interval, CT computed tomography, CXR chest xray, HUS haemolytic uremic syndrome, IL interleukin, IPD invasive pneumococcal disease, LOS length of stay, MIC minimum inhibitory concentration, No number, NP necrotizing pneumonia, NR not reported, OR odds ratio, PCR polymerase chain reaction, PCV7 7-valent pneumococcal conjugate vaccine, PNP pneumococcal necrotizing pneumonia, PPE parapneumonic empyema, PTX pneumothorax, RR risk ratio, SD standard deviation, ST sequence type, US ultrasound, VATS video-assisted thoracoscopy