PNEUMONIAS COMPLICADAS 2017

Prévia do material em texto

ORIGINAL ARTICLE
Surgical Manag
Necrotizing Pn
Jin-Yao Lai*, Wendy
Division of Pediatric Surgery, Dep
Chang Gung University, College o
Received Mar 3, 2016; received in r
Available online - - -
tions were considered uncomplicated.
Results: Thirty-one cases were uncomplicated and 25 were complicated. Operative procedures
syndrome (20% vs.
erative transfusion
p Z 0.02), reopera-
tions (16% vs. 0%; p Z 0.02), and longer postoperative stay (19.8 � 24.2 days vs.
decortications and
ted patients, those
ostoperative stays
), whereas patients
0.23). All patients
e preoperative mor-
ostoperative stays.
* Corresponding author. Division of Pediatric Surgery, Department of Surgery, Chang Gung Memorial Hospital, Linkou, Number 5, Fu-Hsing
Street, Kweishan, Taoyuan 333, Taiwan.
E-mail address: jylai@cgmh.org.tw (J.-Y. Lai).
+ MODEL
http://dx.doi.org/10.1016/j.pedneo.2016.06.002
1875-9572/Copyrightª 2016, TaiwanPediatric Association. Publishedby Elsevier TaiwanLLC. This is an openaccess article under theCCBY-NC-ND
Available online at www.sciencedirect.com
ScienceDirect
journal homepage: http: / /www.pediatr -neonatol .com
Pediatrics and Neonatology (2016) xx, 1e7
Aggressive surgical treatment results in significant clinical improvement. Lobectomy in
11.2 � 5.8 days; p Z 0.03). Four complicated patients, who initially had
limited resections, underwent reoperations. Compared with uncomplica
who underwent decortications and wedge resection required longer p
(23.6 � 9.9 days, p < 0.01 and 21.1 � 30.7 days, p Z 0.04, respectively
who had lobectomy had a similar duration of recovery (9.0 � 2.1 days, pZ
improved significantly at follow-up.
Conclusion: Children with complicated necrotizing pneumonitis have mor
bidities, more major postoperative complications, and require longer p
tracheal intubation (44% vs. 9.7%; p Z 0.008), and hemolytic uremic
3.2%; p Z 0.01). These patients also had higher incidences of intraop
(68% vs. 9.7%; p Z 0.03), major postoperative complications (16% vs. 0%;
included 38 decortications (31 uncomplicated and seven complicated), 14 wedge resections,
and four lobectomies (complicated only). Preoperatively, patients with complicated necro-
tizing pneumonia had a higher incidence of pneumothorax (32% vs. 14.3%; p Z 0.001), endo-
Key Words
decortication;
lobectomy;
necrotizing
pneumonia
Please cite this article in press as: La
Neonatology (2016), http://dx.doi.o
license (http://creativecommons.org/l
ement of Complicated
eumonia in Children
Yang, Yung-Ching Ming
artment of Surgery, Chang Gung Memorial Hospital,
f Medicine, Linkou, Taiwan
evised form May 22, 2016; accepted Jun 12, 2016
Background: There are no well-established indications for the surgical management of acute
necrotizing pneumonitis in children. This study presents our experience regarding this chal-
lenging topic.
Methods: Between 2002 and 2009, 56 necrotizing pneumonitis patients with empyema were
treated surgically. The outcomes were analyzed retrospectively. Computed tomography find-
ings of massive lung necrosis or large cavities involving more than 50% of the involved lobe
were deemed to be complicated necrotizing pneumonitis. Patients without the above indica-
i J-Y, et al., Surgical Management of Complicated Necrotizing Pneumonia in Children, Pediatrics and
rg/10.1016/j.pedneo.2016.06.002
icenses/by-nc-nd/4.0/).
patients with complicated necrotizing pneumonitis may shorten the postoperative course and
ed
he
2.2. Surgical indications and procedures
Indications for surgery included
aggressive medical treatment (ant
pigtail catheterization or chest tu
apy), persistent respiratory dist
and lobectomy (LB).
The 56 patients included 31 (20 men, 11 women) with un-
complicated NP and 25 (12 men, 13 women) with compli-
cated NP. The mean age at presentation was
2 J.-Y. Lai et al
+ MODEL
Please cite this article in press as: La
Neonatology (2016), http://dx.doi.o
persistent fever despite
ibiotics, thoracocentesis,
be, and fibrinolytic ther-
ress, and sepsis. Three
3.46 � 1.98 years. There was no significant difference in
the mean age of patients undergoing DC, WR, or LB. The sex
ratios did not significantly differ between the two groups
(pZ 0.22) or among the three types of surgery (pZ 0.35).
The results are summarized in Tables 1 and 2.
avoid subsequent surgery.
Copyright ª 2016, Taiwan P
open access article under t
by-nc-nd/4.0/).
1. Introduction
Although community-acquired pneumonia generally runs a
benign course, it can be complicated by empyema, pneu-
matocele, necrotizing pneumonia (NP), or lung abscess. NP
is uncommon in the pediatric population and has a very
poor outcome in adults. The incidence of NP has increased
in recent years.1,2
When initial medical treatment of acute NP is unsuc-
cessful in pediatric patients, there are no well-established
indications for surgical management. The result of surgery
is considerably influenced by the extent of necrosis, which
determines the extent of pulmonary resection.3,4 This study
reports the outcomes of surgery for NP in a single
institution.
2. Patients and methods
Between January 2002 and December 2010, 134 patients
with acute NP were admitted to Chang Gung Memorial
Hospital, Linkou, Taiwan. Of these patients, 56 underwent
surgery and were included in this retrospective study. The
following data were collected for each patient: age, sex,
bacterial culture results, surgical technique, number of
admission days prior to the surgery, preoperative place-
ment of a chest tube or pigtail catheter, preoperative
morbidity, transfusion during surgery, length of post-
operative stay (POD), major postoperative complications,
and the incidence of reoperation. All patients were fol-
lowed for >3 years after surgery.
This study was approved by the Ethics Committee of our
hospital (Institutional Review Board 99-0744B). The need
for written consent was waived because of the retrospec-
tive study design.
2.1. Definition of necrotizing pneumonitis
NP was defined as patchy inflammation with microabscesses
and decreased perfusion on computed tomography (CT).5e7
The patients were divided into a Complicated NP group
(with massive lung necrosis or large cavities involving >50%
of the involved lobe) and an Uncomplicated NP group (with
less severe CT findings).
i J-Y, et al., Surgical Management
rg/10.1016/j.pedneo.2016.06.002
All patients received general anesthesia with a single
lumen endotracheal tube and were placed in a lateral
decubitus position. All 38 DCs and five WRs were thor-
acoscopically performed. Two 5-mm ports were placed in
the midaxillary line in the fifth and seventh intercostal
spaces. For DC, pleural fluid was evacuated, and all fibri-
nous debris and loculations were bluntly removed using
forceps under direct thoracoscopic visualization. Samples
were submitted for Gram stain and culture. The chest was
irrigated, and a large-bore chest tube was positioned
under thoracoscopic guidance. For WR, an additional port
was placed in the midaxillary line in the third intercostal
space. Necrotic lung tissue was completely removed, and
the lung abscess was irrigated and evacuated. The lung
defect was repaired using interrupted sutures of 2e0
PROLENE (Polypropylene; Ethicon, Cincinnati, OH, USA).
Lung re-expansion was confirmed at the conclusion of the
procedure.
In nine WRs and all four LBs, posterior thoracotomies
were performed. Fibrinous debris and loculations were
removed. In open WRs, necrotic lung tissues were
completely debrided, lung abscesses were thoroughly irri-
gated and evacuated, and lung defects were repaired using
interrupted sutures with 2e0 PROLENE. In each LB, the
bronchus was closed by double ligation using 2e0 PROLENE,
and several reinforced stitches were placed distal to the
stump. No muscle or pericardial flap was usedto cover the
stump. A large-bore chest tube was positioned under direct
visual guidance.
2.3. Statistical analysis
The mean and standard deviation were calculated for each
group, and Student t test and analysis of variance were
used to detect significant differences between the means,
with p < 0.05.
3. Results
iatric Association. Published by Elsevier Taiwan LLC. This is an
CC BY-NC-ND license (http://creativecommons.org/licenses/
types of operations were performed: thoracoscopic decor-
tication (DC) without lung resection, wedge resection (WR),
of Complicated Necrotizing Pneumonia in Children, Pediatrics and
3.2. Preoperative stay and morbidities
The mean length of preoperative stay was approximately
the same in both groups (11.5 � 6.47 days vs.
13.5 � 9.00 days; p Z 0.16). In the entire series, 44 pa-
tients (78.6%) had preoperative chest tube or pigtail cath-
eter drainage, including 19 with uncomplicated and all 25
with complicated NP (61.3% vs. 100%, p Z 0.0002).
Preoperative fibrinolytic treatment was used in 33 pa-
tients (58.9%), including 16 with uncomplicated and 17 with
complicated NP, which had no significant difference (51.6%
vs. 68%, p Z 0.18).
All patients had significant empyema prior to the sur-
gery. Hemolytic uremic syndrome (HUS) developed in six
patients (10.7%) preoperatively, all of whom had positive
cultures for S. pneumoniae. The occurrence of HUS was
significantly higher in patients with complicated NP (nZ 5,
83.3%) than those with uncomplicated NP (n Z 1, 16.7%;
p Z 0.04). After their renal function stabilized, three pa-
tients with HUS underwent LB (50%; Figure 1), two under-
went WR (33%), and one underwent DC (17%). One patient
who underwent WR required reoperation because of a
Table 1 Results of uncomplicated and complicated
necrotizing pneumonitis (NP) patients.
Uncomplicated
NP
Complicated
NP
p
No. n Z 31 (55.4%) n Z 25 (44.6%)
Sex (M/F) 20/11 12/1 0.22
Age at operation (y) 2.4 � 1.03 4.1 � 1.24 0.39
Preoperative stay (d) 11.5 � 6.47 13.5 � 9.00 0.16
Preoperative chest
tube or pigtail
catheter (%)
61.3 (n Z 19) 100 (n Z 25) 0.0002
Preoperative
fibrinolytic
treatment (%)
51.6 (n Z 16) 68 (n Z 17) 0.18
Preoperative
pneumothorax (%)
14.3 (n Z 1) 32 (n Z 8) 0.001
Preoperative
endotracheal
intubation (%)
9.7 (n Z 3) 44 (n Z 11) 0.008
Preoperative HUS (%) 3.2 (n Z 1) 20 (n Z 5) 0.01
Intraoperative blood 9.7 (n Z 3) 68 (n Z 17) 0.03
Surgical management of necrotizing pneumonia 3
+ MODEL
transfusion (%)
Major postoperative
complications (%)
0 16 (n Z 4) 0.02
Reoperation (%) 0 16 (n Z 4) 0.02
POD (d) 11.2 � 5.8 19.8 � 24.2 0.03
Mortality 0 0
Follow-up (mo) 79.2 � 28.8 66.2 � 28.4 0.09
HUSZ hemolytic uremic syndrome; PODZ postoperative stay.
3.1. Microbiology findings
No child had a known underlying immunological disorder.
Streptococcus pneumoniae was the most common pathogen
(n Z 29, 51.8%), followed by Pseudomonas aeruginosa
(nZ 1, 1.8%) and Staphylococcus aureus (nZ 1, 1.8%). The
remaining 25 patients (44.6%) had no growth on cultures of
pleural effusion or blood.
Table 2 Surgical results in complicated necrotizing pneumoniti
Operation method DC
No. n Z 7 (28
Sex (M/F) 4/3
Age at operation (y) 2.4 � 1.0
Preoperative stay (d) 12.1 � 5.
Preoperative chest tube or pigtail catheter (%) 100 (n Z
Preoperative fibrinolytic treatment (%) 85.7 (n Z
Preoperative pneumothorax (%) 14.3 (n Z
Preoperative endotracheal intubation (%) 42.9 (n Z
Preoperative HUS (%) 0
Intraoperative blood transfusion (%) 42.9 (n Z
Major postoperative complications (%) 28.6 (n Z
Reoperation (%) 28.6 (n Z
POD (d) 23.6 � 9.
Mortality 0
Follow up (mo) 98.6 � 26
DC Z decortication; HUS Z hemolytic uremic syndrome; LB Z lobec
Please cite this article in press as: Lai J-Y, et al., Surgical Management
Neonatology (2016), http://dx.doi.org/10.1016/j.pedneo.2016.06.002
persistent air leak. The mean PODs for the six patients with
HUS were 31 days (DC), 27.5 days (WR), and 8 days (LB),
which did not significantly differ from those for patients
without HUS (18.5 � 34.3 days vs. 14.2 � 16.9 days;
pZ 0.82). The LB group had a shorter POD than the DC and
WR groups and a comparable POD to that of patients with
uncomplicated NP.
Preoperatively, nine patients developed pneumothorax
(16.1%), with the incidence being significantly higher in the
complicated than in the uncomplicated group (32% vs.
14.3%, p Z 0.001). The incidences of preoperative pneu-
mothorax in patients who underwent DC, WR, and LB were
14.3% (n Z 1), 21.4% (n Z 3), and 100% (n Z 4), respec-
tively. The LB group had a significantly higher preoperative
pneumothorax rate than either the DC or the WR group
(p Z 0.000 for both comparisons).
A total of 14 patients were intubated for respiratory
distress prior to the surgery (25%), a rate that was higher in
s patients.
WR LB p
%) n Z 14 (56%) n Z 4 (16%)
7/7 1/3 0.39
3 4.1 � 1.24 3.0 � 1.0 0.44
67 13.8 � 11.21 15.0 � 2.12 0.75
7) 100 (n Z 14) 100 (n Z 4)
6) 78.6 (n Z 11) 0 0.002
1) 21.4 (n Z 3) 100 (n Z 4) 0.000
3) 28.6 (n Z 4) 100 (n Z 4) 0.03
14.3 (n Z 2) 75 (n Z 3) 0.01
3) 71.4 (n Z 10) 100 (n Z 4) 0.03
2) 14.3 (n Z 2) 0 0.28
2) 14.3 (n Z 2) 0 0.28
9 21.1 � 30.7 9.0 � 2.1 0.01
0 0
.0 56.8 � 16.7 42.8 � 2.6 0.08
tomy; POD Z postoperative stay; WR Z wedge resection.
of Complicated Necrotizing Pneumonia in Children, Pediatrics and
4 J.-Y. Lai et al
+ MODEL
the complicated than in the uncomplicated group (44% vs.
9.7%, p Z 0.03). The incidence of preoperative intubation
in the DC, WR, and LB groups was 42.9%, 28.6%, and 100%,
respectively. The LB group had a higher preoperative
intubation rate than either the DC or WR group (p Z 0.004
and p Z 0.03, respectively).
3.3. Operative procedures and reoperation
All 31 patients with uncomplicated NP underwent DC, and
none required reoperation. Of the 25 complicated NP pa-
tients, seven underwent DC (28%), 14 WR (56%), and four LB
(16%); four (16%) required a second operation because of
The only patient undergoing LB as a second procedure
Figure 1 Lobectomy in complicated necrotizing pneumo-
nitis. (A, B) The computed tomographic scan shows a large
cavitation and lung abscess in the left lower lobe in a 3-year-
old girl with hemolytic uremic syndrome. (C) Gangrenous lung
with abscess.
Please cite this article in press as: Lai J-Y, et al., Surgical Management
Neonatology (2016), http://dx.doi.org/10.1016/j.pedneo.2016.06.002
was a 1-month-old girl with staphylococcal NP. The preop-
erative CT showed pneumothorax and extensive lung ne-
crosis without perfusion in the right middle and lower
lobes. Initially, she underwent WR, but a severe air leak,
progressive lung necrosis, and difficulty in weaning from
the ventilator forced a second operation. The remaining
necrotic right upper lobe was resected to prevent air
leakage, resulting in total right pneumonectomy (Figure 3).
Her POD was 120 days. She was breathing well and was no
longer oxygen-dependent at follow-up.
3.4. Complications and postoperative stay
Twenty patients (35.7%) received blood transfusions during
surgery, with a higher rate in the Complicated than in the
Uncomplicated group (68% vs. 9.7%, p Z 0.03). The inci-
dence of transfusion in patients who underwent DC, WR, or
LB was 42.9%, 71.4%, and 100%, respectively. The LB group
had a slightly higher but not significantly different trans-
fusion rate than the DC or WR group (p Z 0.25 and
p Z 0.06, respectively). None of these patients sustained
transfusion-related complications.
Of the 31 patients with uncomplicated NP, only two
(6.8%) had minor air leakages, which spontaneously sub-
sided within 5 days postoperatively. No major complica-
tions were noted in these patients. Of those with
complicated NP, two patients who underwent DC (28.6%)
and two who underwent WR (14.3%)each had recurrent
empyema and a persistent air leak for >2 weeks, and thus
may require reoperation. The major complication rate was
higher in the complicated than in the uncomplicated NP
group (16% vs. 0%; p Z 0.02).
The mean POD in complicated cases was longer than that
in uncomplicated cases (19.8 � 24.2 days vs.
11.2 � 5.8 days, p Z 0.03). Patients with complicated NP
recurrent empyema and persistent air leakage for >2 weeks
(2 after initial WR and 2 after initial DC). The second sur-
geries included three WRs and one LB. No patients who
underwent LB developed any major complication post-
operatively. The PODs of these four reoperated patients
were 28 days, 35 days, 48 days, and 120 days.
Although the reoperation rate for patients who initially
underwent DC and WR was not statistically higher than that
for those whose first surgery was LB (p Z 0.28), the POD
was significantly longer after DC and WR compared with LB
(p Z 0.01 and 0.003, respectively). All four reoperated
patients had required preoperative endotracheal intuba-
tion because of respiratory distress. Three of the four
reoperated patients had pneumothorax preoperatively,
whereas two had HUS.
One patient requiring a second WR was a 2-year-old girl
with complicated NP and HUS. Preoperative CT revealed
extensive lung necrosis with no perfusion in the left lower
lobe (Figure 2). Pneumothorax with a large cavity devel-
oped. WR was performed, but severe air leakage persisted
postoperatively. Two weeks after the first surgery, she un-
derwent a second WR because of the air leak and respira-
tory distress. The air leak persisted for another month after
the second procedure. Her POD was 48 days prior to com-
plete recovery.
of Complicated Necrotizing Pneumonia in Children, Pediatrics and
Surgical management of necrotizing pneumonia 5
+ MODEL
who underwent DC or WR required a longer postoperative
stay than those with uncomplicated NP (23.6 � 9.9 days vs.
11.2 � 5.8 days, p Z 0.0001 and 21.1 � 30.7 days vs.
11.2 � 5.8 days, p Z 0.04, respectively), most probably
because of the higher incidence of postoperative compli-
cations and reoperation. However, patients with compli-
cated NP who underwent LB as an initial procedure had
PODs similar to those with uncomplicated NP
(9.0 � 2.1 days vs. 11.2 � 5.8 days, pZ 0.23). The PODs of
DC and WR are not significantly changed after exclusion of
the reoperative patients. In the DC group, the PODs prior to
and after reoperative patient exclusion were
23.6 � 9.9 days versus 21.8 � 10.4 days (p Z 0.39). In the
WR group, the PODs prior to and after outlier exclusion
were 21.1 � 30.7 days versus 10.7 � 5.0 days (p Z 0.13).
There was no operative mortality in this series.
3.5. Follow-up
All patients were followed for at least 3 years (mean,
74.4 � 29.1 months). All but one patient had normal or
significantly improved chest X-rays at follow-up. No evi-
dence of cavitary lesions or pulmonary fibrosis was detec-
ted. Only one patient lacked a right lung owing to
pneumonectomy. Pulmonary function tests were not per-
formed in any of the patients. The growth and development
of all 56 patients were within normal limits.
Figure 2 Wedge resection in a patient with complicated necrotiz
area of lung necrosis, and a small cavity occupies the left lower lob
air leakage 2 weeks after the first operation and 2 days prior to th
second operation.
Please cite this article in press as: Lai J-Y, et al., Surgical Management
Neonatology (2016), http://dx.doi.org/10.1016/j.pedneo.2016.06.002
4. Discussion
Necrotizing pneumonitis constitutes a complication of se-
vere pneumonia, with a dominant area of necrosis con-
taining a variable number of thin-walled cavities. The
mechanism involves thrombotic occlusion of the alveolar
capillaries associated with adjacent inflammation, result-
ing in ischemia, and eventually necrosis of the lung pa-
renchyma. Most reported cases of NP have been in adults
and are usually associated with staphylococcal pneu-
monia.5,6 In pediatric patients, the most common pathogen
is S. pneumoniae, followed by S. aureus and Klebsiella
pneumoniae.5,6,8 Clinically, the severity of NP is substan-
tially influenced by the degree of inflammation, area of
necrosis, time course, and degree of sepsis.
Preoperative imaging studies are critical in treating NP.
In our series, all children underwent chest CT at least once.
The benefits included estimation of the degree of paren-
chymal destruction, detection of parenchymal and pleural
complications, detection of occult changes (such as
microabscesses), and delineation of areas without perfu-
sion.5,6,8,9 These findings can identify areas of extensive
local necrosis and demonstrate the failure of medical
management. As the infection progresses, necrotic areas
coalesce to form a single or a few large cavities containing
sloughed lung tissue that floats atop pus collected in the
bottom of the abscess. In the final stage, a large mass of
ing pneumonitis with reoperation. (A, B) Pneumothorax, a large
e. (C) Improving necrotizing pneumonitis with severe persistent
e second operation. (D) Chest X-ray scan 28 months after the
of Complicated Necrotizing Pneumonia in Children, Pediatrics and
6 J.-Y. Lai et al
+ MODEL
necrotic lung tissue can cause respiratory distress and
pneumothorax. Some investigators consider central necro-
sis affecting >50% of the involved lobe to be gangrene.9 In
our series, a cavitary lesion occupying >50% of the affected
lobe was defined as complicated NP.
All of our patients had received complete standard
medical therapy before they were referred for surgery.
Common preoperative morbidities included pleural empy-
ema (100%), preoperative endotracheal intubation for res-
piratory distress (26.8%), pneumothorax (12.5%), and HUS
(10.7%), demonstrating the severity of this disease. Under
these circumstances, it is essential to focus on controlling
sepsis. The devitalized parts of the lung or massive ab-
scesses are difficult to remove with simple drainage alone.
In our opinion, traditional surgical drainage operations10 to
unroof an abscess cavity and debride dead tissue are not
sufficient to treat very complicated NP. If the abscess
cannot be effectively removed, a severe air leak may
result.
The optimal surgical treatment for acute NP with em-
pyema remains controversial. Some investigators recom-
mend formal LB for most cases,3,4,8,11 whereas others
suggest that LB is rarely necessary and prefer to perform DC
to preserve lung parenchyma.12,13 Because the severity of
NP is considerably influenced by the degree of necrosis, the
treatment should be based on the severity of destruction
Figure 3 Wedge resection in complicated necrotizing pneumoniti
perfusion in the right middle and lower lobes with pneumothorax
evident perfusion after the first procedure (wedge resection). (C) O
the second procedure (lobectomy). (D) Chest X-ray scan 13 months
had total right pneumonectomy.
Please cite this article in press as: Lai J-Y, et al., Surgical Management
Neonatology (2016), http://dx.doi.org/10.1016/j.pedneo.2016.06.002
and any associated complications. For uncomplicated NP, a
more conservative operation (i.e., DC) is usually sufficient.
The goal of surgery is adequate drainage of the empyema.
The lung infection is treated with antibiotics.
In more complicated and devastating NP, aggressive
surgical management involving lung resection can be
effective and is associated with less postoperative
morbidity, as observed in our series. This type of surgery
involves clearing the pleural infection (DC) and removing
the necrotic lung tissue by WR or LB to avoid further com-
plications. In the present series, LB was performed in five
patients (in one case as a secondary salvage procedure).
Although the requirement for intraoperative blood trans-
fusion was significantly higher in complicatedthan in un-
complicated NP cases, there were no major transfusion-
related complications. The need for a transfusion illus-
trates the presence of severe tissue inflammation and the
difficult surgical techniques required when managing
complicated NP. All four patients undergoing primary LB
were discharged in approximately 9 days without post-
operative morbidity. In contrast, both DC and WR in pa-
tients with complicated NP were associated with a slightly
higher incidence of reoperation and significantly longer
postoperative stay. These findings demonstrate that inad-
equate surgical removal of necrotic tissue can lead to more
severe morbidity in complicated cases. Formal LB might be
s requring pneumonectomy. (A) Extensive lung necrosis without
. (B) A large cavity in the remaining right upper lobe without
nly a small piece of necrotic right upper lobe detected during
after the second operation. After two procedures, the patient
of Complicated Necrotizing Pneumonia in Children, Pediatrics and
a better procedure in highly selected patients. In the pre-
sent series, the risk factors for reoperation were preoper-
ative endotracheal intubation for respiratory distress,
preoperative pneumothorax, and HUS.
In devastating NP, death has been reported following
LB.3,4,14 Schweigert et al4 reported a 15.9% mortality rate
for pneumonectomy and LB in patients with NP with lung
gangrene. In their series, two patients died after under-
going pneumonectomy (n Z 11) and five died following LB
Conflicts of interest
The authors have no conflicts of interest relevant to this
article.
References
1. Hsieh YC, Hsiao CH, Tsao PN, Wang JY, Hsueh PR, Chiang BL,
Surgical management of necrotizing pneumonia 7
+ MODEL
(n Z 26), whereas all who underwent segmentectomy
(n Z 7) survived. The patients in that series were adults
with pneumonia complicated by sepsis. Pediatric patients
typically have healthier lungs than adults; therefore,
children have a better recovery after aggressive manage-
ment of NP. One of our patients, after an initial WR and a
second LB, had total loss of the right lung. She survived her
NP and was not oxygen-dependent 44 months after
surgery.
None of the children in our series had an identifiable
comorbidity, such as congenital lung lesions, cystic fibrosis,
bronchiectasis, neurologic compromise, or immunodefi-
ciency, all of which may increase the risk of complicated
pneumonia. However, six patients developed HUS preop-
eratively. An increasingly recognized cause of atypical HUS
is invasive S. pneumoniae infection.15 Most of such cases
progress to acute renal failure within a few days after the
onset of disease. The relationship between NP and HUS
remains unclear.1,16 All of the patients with HUS in our se-
ries underwent hemodialysis and experienced complete
recovery of renal function in 2e3 weeks. Lung destruction
was severe in five of the six patients with HUS. They all
required lung resection, including three LBs (50%) and two
WRs (33%). The only patient with uncomplicated CT findings
underwent DC. Two of the six patients with HUS (33%) who
underwent WRs required second operations. All patients
survived with normal renal function and satisfactory res-
piratory function after follow-up for >3 years. Although the
numbers were small, the patients with HUS undergoing LB
appeared to have a shorter POD than those treated with DC
or WR. This finding stresses the role of aggressive surgical
management in NP complicated with HUS.
In conclusion, more than half of the NP cases in our se-
ries were caused by S. pneumoniae. Preoperative CT
enabled excellent evaluation of the complex anatomy.
Children with complicated NP had more preoperative
morbidities and major postoperative complications and
required longer PODs than those with uncomplicated NP.
Inadequate surgical treatment may lead to a longer POD
and reoperation. Patients with massive pulmonary necrosis,
preoperative intubation for respiratory distress, pneumo-
thorax, or HUS may require formal LB, which usually
shortens the postoperative course, avoids a second opera-
tion, and has a satisfactory long-term outcome.
Please cite this article in press as: Lai J-Y, et al., Surgical Management
Neonatology (2016), http://dx.doi.org/10.1016/j.pedneo.2016.06.002
et al. Necrotizing pneumococcal pneumonia in children: the
role of pulmonary gangrene. Pediatr Pulmonol 2006;41:623e9.
2. Sawicki GS, Lu FL, Valim C, Cleveland RH, Colin AA. Necrotising
pneumonia is an increasingly detected complication of pneu-
monia in children. Eur Respir J 2008;31:1285e91.
3. Schweigert M, Dubecz A, Beron M, Ofner D, Stein HJ. Surgical
therapy for necrotizing pneumonia and lung gangrene. Thorac
Cardiovasc Surg 2013;61:636e41.
4. Schweigert M, Giraldo Ospina CF, Solymosi N, Karmy-Jones R,
Dubecz A, Ferna´ndez MJ, et al. Emergent pneumonectomy for
lung gangrene: does the outcome warrant the procedure? Ann
Thorac Surg 2014;98:265e70.
5. Donnelly LF, Klosterman LA. Pneumonia in children: decreased
parenchymal contrast enhancement-CT sign of intense illness
and impending cavitary necrosis. Radiology 1997;205:817e20.
6. Donnelly LF, Klosterman LA. Cavitary necrosis complicating
pneumonia in children: sequential findings on chest radiog-
raphy. AJR Am J Roentgenol 1998;171:253e6.
7. Penner C, Maycher B, Long R. Pulmonary gangrene. A compli-
cation of bacterial pneumonia. Chest 1994;105:567e73.
8. Cowles RA, Lelli Jr JL, Takayasu J, Coran AG. Lung resection in
infants and children with pulmonary infections refractory to
medical therapy. J Pediatr Surg 2002;37:643e7.
9. Reimel BA, Krishnadasen B, Cuschieri J, Klein MB, Gross J,
Karmy-Jones R. Surgical management of acute necrotizing lung
infections. Can Respir J 2006;13:369e73.
10. Postma MH, le Roux BT. The place of external drainage in the
management of lung abscess. S Afr J Surg 1986;24:156e8.
11. Ayed AK, Al-Rowayeh A. Lung resection in children for infec-
tious pulmonary diseases. Pediatr Surg Int 2005;21:604e8.
12. Kalfa N, Allal H, Lopez M, Counil FO, Forgues D, Guibal MP,
et al. An early thoracoscopic approach in necrotizing pneu-
monia in children: a report of three cases. J Laparoendosc Adv
Surg Tech A 2005;15:18e22.
13. Subramaniam R, Joseph VT, Tan GM, Goh A, Chay OM. Experi-
ence with video-assisted thoracoscopic surgery in the man-
agement of complicated pneumonia in children. J Pediatr Surg
2001;36:316e9.
14. Al-Saleh S, Grasemann H, Cox P. Necrotizing pneumonia
complicated by early and late pneumatoceles. Can Respir J
2008;15:129e32.
15. Brandt J, Wong C, Mihm S, Roberts J, Smith J, Brewer E, et al.
Invasive pneumococcal disease and hemolytic uremic syn-
drome. Pediatrics 2002;110:371e6.
16. Lee CF, Liu SC, Lue KH, Chen JP, Sheu JN. Pneumococcal
pneumonia with empyema and hemolytic uremic syndrome in
children: report of three cases. J Microbiol Immunol Infect
2006;39:348e52.
of Complicated Necrotizing Pneumonia in Children, Pediatrics and
	Surgical Management of Complicated Necrotizing Pneumonia in Children
	1. Introduction
	2. Patients and methods
	2.1. Definition of necrotizing pneumonitis
	2.2. Surgical indications and procedures
	2.3. Statistical analysis
	3. Results
	3.1. Microbiology findings
	3.2. Preoperative stay and morbidities
	3.3. Operative procedures and reoperation
	3.4. Complications and postoperative stay
	3.5. Follow-up
	4. Discussion
	Conflicts of interest
	References