Comparison of Urokinase and Video-assisted
Thoracoscopic Surgery for Treatment of
Childhood Empyema
Samatha Sonnappa, Gordon Cohen, Catherine M. Owens, Carin van Doorn, John Cairns, Sanja Stanojevic,
Martin J. Elliott, and Adam Jaffé
Department of Respiratory Medicine, Department of Cardio-Thoracic Surgery, and Department of Radiology, Great Ormond Street Hospital
for Children NHS Trust; Portex Anaesthesia, Intensive Therapy and Respiratory Unit, Institute of Child Health; and Department of Public
Health and Policy, London School of Hygiene and Tropical Medicine, London, United Kingdom
Background: Despite increasing incidence and morbidity, little evi-
dence exists to inform the best management approach in childhood
empyema.
Aim: To compare chest drain with intrapleural urokinase and pri-
mary video-assisted thoracoscopic surgery (VATS) for the treatment
of childhood empyema.
Methods: Children were prospectively randomized to receive either
percutaneous chest drain with intrapleural urokinase or primary
VATS. The primary outcome was the number of hospital days after
intervention. Secondary end points were number of chest drain
days, total hospital stay, failure rate, radiologic outcome at 6 mo,
and total treatment costs.
Results: Sixty children were recruited. The two groups were well
matched for demographics; baseline characteristics; and hemato-
logic, biochemical, and bacteriologic parameters. No significant
difference was found in length of hospital stay after intervention
between the two groups: VATS (median [range], 6 [3–16] d) versus
urokinase (6 [4–25] d) (p ⫽ 0.311; 95% confidence interval, ⴚ2 to 1).
No difference was demonstrated in total hospital stay: VATS versus
urokinase (8 [4–17] d and 7 [4–25] d) (p ⫽ 0.645); failure rate: 5
(16.6%); and radiologic outcome at 6 mo after intervention in both
groups. The mean (median) treatment costs of patients in the uroki-
nase arm $9,127 ($6,914) were significantly lower than those for
the VATS arm $11,379 ($10,146) (p ⬍ 0.001).
Conclusions: There is no difference in clinical outcome between in-
trapleural urokinase and VATS for the treatment of childhood empy-
ema. Urokinase is a more economic treatment option compared
with VATS and should be the primary treatment of choice. This study
provides an evidence base to guide the management of childhood
empyema.
Keywords: intrapleural urokinase; primary video-assisted thoracoscopic
surgery; prospective randomized trial
The incidence of empyema is increasing worldwide, causing sig-
nificant childhood morbidity with an estimated 0.6% of child-
hood pneumonia progressing to empyema (1–4). The aim of
treatment in empyema is to sterilize the pleural cavity, reduce
fever, and ensure full expansion of the lung and return to normal
function. Many treatment options are available, including antibi-
otics alone or in combination with thoracocentesis, chest-drain
(Received in original form January 6, 2006; accepted in final form April 28, 2006 )
Research at the Institute of Child Health and Great Ormond Street Hospital for
Children NHS Trust benefits from Research and Development funding received
from the NHS Executive.
Correspondence and requests for reprints should be addressed to Samatha
Sonnappa, M.D., D.Ch., M.R.C.P., F.R.C.P.Ch., Portex Anaesthesia, Intensive
Therapy and Respiratory Unit, Level 6, Cardiac Wing, Institute of Child Health,
30, Guilford Street, London WC1N 1EH, UK. E-mail: s.sonnappa@ich.ucl.ac.uk
Am J Respir Crit Care Med Vol 174. pp 221–227, 2006
Originally Published in Press as DOI: 10.1164/rccm.200601-027OC on May 4, 2006
Internet address: www.atsjournals.org
insertion, chest drain and fibrinolytics, mini-thoracotomy, open
decortication, and video-assisted thoracoscopic surgery (VATS).
However, treatment is not standardized and currently patient
care is dependent on local practice and physician preference.
These issues were recently highlighted by the British Thoracic
Society (3) (www.brit-thoracic.org.uk) in the publication of
guidelines on the management of pleural infection in children,
which states that there is a lack of grade A evidence available
to inform the best management approach (5). Urokinase has
been demonstrated as safe in several uncontrolled studies, and
has been found to play an important role in the treatment of
empyema (6–8). Some centers use fibrinolytics such as urokinase,
and in the event of failure patients will undergo either open
decortication or VATS (9). In a large controlled trial, administra-
tion of intrapleural streptokinase in adults with empyema was
not found to reduce mortality, rate of surgery, or length of
hospital stay (10). However, in the only randomized prospective
study in children comparing the instillation of urokinase through
percutaneous chest drains to normal saline, the urokinase group
had a significantly reduced hospital stay after intervention
(7.4 d vs. 9.5 d; ratio of geometric means 1.28; confidence interval
[CI], 1.16–1.41) (11). This study also found shorter hospital stay
with small percutaneous chest drain insertion compared with
large bore chest drains (7.2 d vs. 9.4 d). Conversely, the only
other randomized controlled study of fibrinolytics that compared
instillation of streptokinase with normal saline in Indian children
found no difference between all outcome measures in both
groups (12).
Interest in using primary VATS in the treatment of empyema
in children has increased over the past few years (13). Kern
and Rodgers were the first to use VATS for the treatment of
empyema in children in 1993 (14). Since then there have been
various case series reports of its success in children (14–23).
Subramaniam and coworkers reported a postoperative hospital
stay of 4.63 d ⫾ 0.33 d (20), and Grewal and colleagues reported
one of 4.9 d ⫾ 2.7 d (24). The single randomized controlled
study comparing fibrinolytics (streptokinase) to VATS in adults
found that the VATS group had a higher primary success, fewer
days in hospital, and fewer days with a chest drain (25). There
has been no such study in an equivalent pediatric group, and
adult data cannot be extrapolated to children as empyema has
an estimated mortality of 20% in adults (26), whereas in children
death rarely occurs. The only study in children to compare uroki-
nase and early VATS was a retrospective review of clinical
practice which showed that patients who received VATS had a
significant reduction in total length of hospital stay (15).
Our study is the first randomized prospective trial to compare
chest drain with intrapleural urokinase against primary VATS
for the treatment of empyema in children.
Some of the results of this study have been previously re-
ported in the form of an abstract (27).
222 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 174 2006
Hypothesis
We hypothesized that patients with empyema treated with
primary VATS had a shorter postoperative hospital stay than
patients receiving percutaneous chest drain insertion with in-
trapleural urokinase.
METHODS
Study Design
This study was a prospective, randomized trial performed over 3 yr at
Great Ormond Street Hospital for Children (GOSH) (London, UK).
The study subjects were recruited from patients referred to GOSH, a
tertiary respiratory center, for further management of parapneumonic
effusions.
Patients were included if they were under the age of 16 yr and had
radiographic evidence of empyema (i.e., pleural fluid on chest X-ray
[CXR] and ultrasound). Indications for drainage were a persistent fever
of 38⬚C (100⬚F) or greater after more than 24 h of parenteral antibiotic
treatment or respiratory distress (tachypnea and/or oxygen require-
ment) caused by the pleural collection. Patients were excluded from
the study if thoracocentesis or chest drain insertion was performed or
attempted at the referring hospital, if there was a history of an underly-
ing cardiac disease or previous cardiac surgery or known immunodefi-
ciency. Each patient had blood culture, hemoglobin, white blood cell
(WBC) and differential counts, platelet count, coagulation profile,
C-reactive protein (CRP), electrolytes, albumin, lactate dehydrogenase
(LDH), CXR, and chest ultrasound scan (USS) performed on entry
into the study. Those with low platelets or abnormal clotting received
VATS even if assigned to the urokinase arm, and were analyzed on
the basis of intention to treat. Empyema was categorized into three
stages based on the appearance of pleural fluid on USS: Stage 1, an-
echoic nonseptated fluid; Stage 2, hyperechoic fluid with fibrinous septa-
tion; and Stage 3, hyperechoic loculations with or without thick parietal
peel (28, 29).
Following the intervention pleural fluid was sent for culture, broad
range16S rDNA polymerase chain reaction (PCR), WBC, albumin, and
LDH. Broad range 16S rDNA PCR was performed on pleural fluid
using a previously described technique (30).
Randomization
The randomization scheme was generated by using the internet web
site http://www.randomization.com. The generator randomizes each
subject to a single treatment by using random allocation. Written in-
formed consent was obtained from parents (and patients, where applica-
ble) before patients were randomly allocated to one of two treatment
arms (i.e., urokinase or VATS arm). The randomization code was held
by the trial coordinator (A.J.) during working hours, and during out
of hours the attending pediatrician had access to the code. The trial
coordinator was contacted by telephone to reveal treatment allocation.
Treatment Protocol
Patients were randomized to receive either percutaneous chest drain
with intrapleural urokinase or VATS. The percutaneous chest drain
(Thal-Quick 8 or 10 Fr; William Cook Europe, Bjaererskov, Denmark)
for the urokinase arm was inserted using the Seldinger technique in a
space marked by chest USS, under general anesthesia. Pleural fluid
was allowed to drain out first, after which intrapleural urokinase was
instilled every 12 h for 3 d, in a dose of 10,000 U in 10 ml normal saline
in children under 1 yr of age and 40,000 U in 40 ml normal saline in
children above 1 yr of age as previously described (11). After instillation
the drain was clamped for 4 h and mobilization of the patient was
encouraged. The drain was then unclamped and placed on negative
suction pressure of 10–20 cm of H2O until the next instillation. Failure
was defined as persistent fever ⭓ 38.0⬚C (100⬚F), 4 d after intervention
associated with persistence of fluid on pleural USS. Those who failed
urokinase treatment had secondary VATS.
VATS was performed under general anesthesia with either one or
both lungs ventilated, depending on the size of the child. Two or three
ports were made in the chest, with the child in the lateral decubitus
position. One port was used for the camera and the others for grasping
instruments. The chest cavity was insufflated with carbon dioxide to
aid collapse of the lung for better visualization. The free fluid was
evacuated and loculations drained, the fibrinous adhesions were sepa-
rated, and the pleural debris removed from the pleural lining using
endoscopic grasping forceps or by extensive irrigation and suction. If
on inspecting the pleural cavity and trial of thoracoscopic decortication
the surgeon deemed VATS to be inappropriate (thick peel preventing
lung expansion), the procedure was converted to a mini-thoracotomy.
This was deemed as a failure of VATS for our trial purpose. After the
procedure, one or two chest drains were placed in the portholes on
negative suction pressure of 10–20 cm of H2O to facilitate further
drainage. Adverse events related to treatment were recorded.
In both treatment groups the chest drains were removed when there
was minimal drainage (40–60ml/24 h), and the patients were discharged
home if they remained afebrile for 24 h after drain removal and at
the attending pediatrician’s discretion. Duration of hospital stay after
intervention was calculated from the date of procedure to the date of
discharge. All patients initially received intravenous cefuroxime and
oral azithromycin, which were subsequently rationalized according to
microbiological results.
Outcome Measures
The main outcome studied was the number of days in hospital, after
intervention. Secondary end points were number of days with chest
drainage, total hospital stay, failure rate of assigned treatment, adverse
events, CXR changes at 6 mo after intervention, and total costs of each
treatment modality.
CXRs (postero-anterior views) were performed at 6 mo after inter-
vention. CXRs were scored in separate sessions by two observers (radi-
ologists at different levels of seniority) blinded to treatment and out-
come parameters. The CXRs were assessed for the following five
parameters: pleural thickening, parenchymal changes, bronchial wall
dilatation and thickening, lung expansion, and lung attenuation. If any
one of these parameters were found to be abnormal, then the CXRs
were classified as abnormal. Interobserver reliability was calculated
using a Kappa statistic.
Sample Size and Data Analysis
We believed that a difference in length of post-intervention hospital
stay of 2 d between the two treatment arms would be clinically impor-
tant. This rationale was based on previous VATS case series reported
by Subramaniam and colleagues (20) and Grewal and coworkers (24)
in which the mean postoperative stay was 4.63 d ⫾ 0.33 d and 4.9 d ⫾
2.7 d, respectively, versus 7.39 d (ratio of geometric means 1.28) re-
ported in the urokinase study by Thomson and colleagues (11). To
demonstrate this difference with 5% significance and 80% power, a
minimum of 29 patients were needed in each study group.
Baseline characteristics were compared between the two groups
using ␹2, t tests, and Mann-Whitney U tests where appropriate. All data
were analyzed according to the intention-to-treat principle. A p value
of ⭐ 0.05 was considered statistically significant.
Cost Analysis
Variations in resources use by patients were captured by recording the
procedures each received and their length of stay. The cost analysis
was based on a simple economic model. The cost of patients randomized
to VATS depends on the cost of VATS, the cost of VATS followed
by open surgery, and the proportion going on to open surgery. The
cost of treatment for those randomized to urokinase depends on the
cost of urokinase treatment, the cost of urokinase followed by surgery,
and the proportions receiving VATS and undergoing open surgery.
Individual patient costs were estimated by multiplying these data by
the appropriate unit cost. All patients undergoing VATS were assumed
to have a computed tomography (CT) scan, which reflects the clinical
practice of the surgeons. It is not our usual practice to request a CT
scan for patients undergoing urokinase. However, all subjects received
a CT scan as part of another study examining the utility of CT scanning
in management of childhood empyema. The costs were analyzed both
with the inclusion and exclusion of CT scanning in the urokinase group.
Patients differed in terms of the dose of urokinase received (depending
on their age). All unit cost data came from GOSH and were in terms
Sonnappa, Cohen, Owens, et al.: Urokinase and VATS in Childhood Empyema
of 2005 prices (Table 1). The costs of the two treatment strategies
were then compared using nonparametric methods (Mann-Whitney,
Kolmlgorov-Smirnov, and Wald-Wolfowitz tests).
Ethical Aspects
The project was approved by the local ethics committee. The trial is
fully registered with clinicaltrials.gov (ID: NCT00144950).
RESULTS
Patients
From January 2002 to February 2005, 80 children were referred
to GOSH for further management of pleural effusions from
secondary care centers. Of these, 60 patients were randomized,
30 into each treatment arm (Figure 1). Of the 30 patients random-
ized to VATS, 25 had VATS only, 4 had VATS and mini-
thoracotomy due to the presence of thick peel preventing full
lung expansion, and 1 patient had VATS performed twice. Of
the 30 patients randomized to the urokinase arm, 28 had uroki-
nase and 2 had VATS, as they had prolonged clotting. The
two groups were well matched for demographics and baseline
characteristics including duration of illness before intervention,
hematologic, biochemical, and bacteriologic parameters (Table 2).
Primary End Point
We did not find a clinically significant difference in hospital stay
after intervention between the two treatment groups (p ⫽ 0.311,
95% CI of the median difference, ⫺2 to 1). Patients with VATS
had a median post-intervention hospital stay of 6 d (range, 3–16 d)
compared with children who received urokinase (6 d; range,
4–25 d).
Secondary End Points
When the number of chest drain in situ days were compared in
both groups, we found that chest drains were removed 1 d earlier
in the VATS group, which was of borderline statistical difference
(p ⫽ 0.055). However, this was not clinically significant, as it did
not have an impact on the hospital stay after intervention. There
was no difference in the total inpatient days between the VATS
and urokinase groups (median [range] 8 [4–17] d vs. 7 [4–25] d;
p ⫽ 0.645). Failure rates were similar in both treatment groups.
Four patients in the VATS group had the procedure converted
to a mini-thoracotomy, due to a thick peel preventing lung expan-
sion, and one patient had VATS performed twice. Five patients
in the urokinase arm failed to respond and had to proceed to
have either VATS or mini-thoracotomy. Four of these five fail-
ures were from “Winter 2003.”
A total of 16 patients were lost to follow-up at 6 mo, and
CXRs at 6 mo after intervention were performed in 44 patients
(24 in the VATS arm and 20 in the urokinase arm). Thirty-nine
(88.6%) of the patients had abnormal CXRs (21 in the VATS
arm and 18 in the urokinase arm). There was moderate agree-
ment between the two radiologists in classifying the CXRs as
TABLE 1. INDIVIDUAL COSTS FOR VATS AND UROKINASE
Unit Costs VATS Intrapleural Urokinase
Theatre $2,030 $712
Surgeon $1,506 $300
Anesthetist $585 $180
Urokinase NA $330 or $639 depending on age of child
CT Chest $864 NA
Ward cost per day $781 $781
Definition of abbreviations: CT chest ⫽ computed tomography chest scan;
VATS ⫽ video-assisted thoracoscopic surgery.
223
normal or abnormal using our devised scoring system (Kappa ⫽
0.54). There was no significant difference in the radiologic out-
come between the two groups (p ⫽ 0.27).
When assessed for USS staging at entry into the study 16 (10
in the urokinase group and 6 in the VATS group) had Stage 1
disease, 21 (8 in the urokinase group and 13 in the VATS group)
had Stage 2 disease, and 23 (12 in the urokinase group and 11
in the VATS group) had Stage 3 disease. When the treatment
groups were compared depending on USS staging, the VATS
group had a reduced hospital stay of borderline statistical sig-
nificance noted in the USS Stage 1 patients: VATS versus uroki-
nase (5 [3–5] d vs. 6 [4–25] d; p ⫽ 0.056). However, this result
is likely to be skewed by an outlier in the urokinase group who
had a hospital stay of 25 d. When this outlier was omitted from
analysis, the effect disappeared (p ⫽ 0.088). We did not find a
difference in postintervention hospital stay between the two
groups in USS Stage 2: VATS versus urokinase (6 [3–16] d vs.
5 [5–9] d; p ⫽ 0.86) and USS Stage 3 (7 [4–12] d vs. 7 [5–19]
days; p ⫽ 0.83).
Adverse Events
We did not have any adverse events directly related to the
treatment in either group. However, in the urokinase group,
chest drains fell out in four patients, requiring reinsertion and
therefore prolonging hospital stay. Complications not directly
related to the treatment included pyohemothorax post–chicken
pox infection in a patient from the VATS group, lung abscess
in four patients (three in the VATS group), hemolytic uremic
syndrome in two patients (both in VATS group), and acute
glomerulonephritis in one (urokinase group).
Cost Analysis
The costs in the VATS arm exceeded those in the urokinase
arm by 25%. This was confirmed by the median test, which
rejected the null hypothesis of the equality of the distributions at
p ⬍ 0.001 (exact significance). The Mann-Whitney, Kolmlgorov-
Smirnov, and Wald-Wolfowitz tests all strongly suggest that a
higher cost is associated with VATS when compared with uroki-
nase (p ⬍ 0.001). The mean (median) treatment costs of patients
in the urokinase arm ($9,127 [$6,914]) were lower compared
with those in the VATS arm ($11,379 [$10,146]). Therefore,
VATS on average is $2,250 more expensive for each patient
than is intrapleural urokinase. If the costs of CT chest scans
were excluded from analysis, the mean (median) treatment costs
of patients in the VATS arm were still significantly higher
($10,515 [$9,282]) compared with the urokinase arm ($8,955
[$6,913]) (p ⬍ 0.001).
Microbiology
Seven (23%) patients in each of the groups had positive blood
or pleural fluid cultures. Forty-seven (78%) of the pleural fluid
samples were analyzed using broad range 16S rDNA PCR and
culture. Of these 27 (45%) were positive for fully penicillin-
sensitive Streptococcus pneumoniae. When all microbiological
data were included, there was no significant difference in caus-
ative organisms isolated between the two treatment groups
(p ⫽ 0.752), and further analysis demonstrated that there was
no impact of the causative organism on the length of hospital
stay in both groups.
DISCUSSION
This prospective study is the first to demonstrate that there is
no significant clinical difference in duration of hospital stay after
intervention between percutaneous chest drain with intrapleural
urokinase and primary VATS for the treatment of empyema in
224 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 174 2006
children. While there have been many case series comparing
surgical interventions with fibrinolytics, none were properly con-
trolled randomized studies. The aim of instillation of fibrinolytics
into the pleural cavity is to lyse the fibrinous strands and clear
lymphatic pores, thus improving drainage (11). There have been
more than 10 published reports on the use of fibrinolytics in
children (6, 7, 12, 31–42), but only two are randomized controlled
trials, neither of which compared fibrinolytics to surgical treat-
ment. The case series reports describe management of empyema
in more than 300 children using streptokinase, urokinase, al-
teplase, or tissue plasminogen activator, but each used different
protocols and hence is not comparable. In these series, the suc-
cess rates were approximately 80–90%, and it was clearly demon-
strated that the use of fibrinolytics is safe in children. One surgical
concern is that patients may be more likely to fail rescue VATS
treatment following urokinase, as it is suggested that urokinase
causes intrapleural loculations to become very adhesive, and
increases the difficulty of the VATS procedure (43, 44), although
this may simply be a result of operating at a later stage.
VATS for the primary treatment of empyema in children has
been gaining popularity over the past decade. Proponents of
VATS suggest that it has the potential advantage over open
surgery of limiting the morbidity to skin, muscles, nerves, and
supporting structures that occurs after a large surgical incision
(13) and that entails pain, infection, limitation of movement,
and cosmetic scarring (45). Furthermore, VATS may reduce
cytokine responses compared with conventional surgery (46).
However, these statements are based on clinical experience
Figure 1. CONSORT flow diagram showing progress
through phases of the trial.
rather than carefully conducted trials. The major limitation of
VATS is that it is highly dependent on the skill of the operator;
poor results in some centers have been reported (16), and surgi-
cal expertise to perform pediatric VATS is limited to a few major
centers in the United Kingdom (47, 48).
Our study was not designed as an equivalence study, as we
based our hypothesis on previously reported results and ex-
pected patients randomized to VATS to have a shorter hospital
stay. We chose our primary outcome measure as length of hospi-
tal stay after intervention rather than as total hospital stay, as
we were evaluating the effect of the two different management
approaches. The two treatment groups in the study were well
matched for baseline characteristics. We did not find a difference
in hospital stay after intervention between the two groups. The
precision of the estimates (95% CI) suggests that the true differ-
ence in the hospital stay between the two groups ranges from
two fewer days to one extra day for the VATS group. This
suggests that there is no clinically significant difference of more
than two hospital days between these two treatment groups.
Alternatively, an equivalence study to determine no difference
between the two treatments with a precision of less than 1 d
would have required over 1,000 patients. Even if a multicentered
trial was conducted, recruiting such a large number of patients
into the study would be logistically difficult and is unlikely ever
to be performed, particularly as pediatric VATS expertise is
limited to a few specialist centers in the United Kingdom.
It has been suggested that ultrasound is useful to stage the
disease in childhood empyema (28). We did not find USS staging
Sonnappa, Cohen, Owens, et al.: Urokinase and VATS in Childhood Empyema 225

TABLE 2. BASELINE CHARACTERISTICS OF THE PATIENTS ACCORDING TO STUDY GROUP

Variable VATS (n ⫽ 30) Urokinase (n ⫽ 30) p Value

Sex, no.
Females 14 13 0.79
Males 16 17
Age, yr
Median 3.57 3.07 0.355
Interquartile range 2.28–7 2.28–5.38
Illness days before admission 10 (6–34) 9 (2–37) 0.458
Illness days before intervention 11 (6–37) 9 (5–38) 0.263
Pre-op days after admission 1 (0–3) 1 (0–3) 0.071
WBC ⫻ 109/L
Median 18 15.22 0.223
Interquartile range 10.8–23.3 10.6–20.4
CRP, mg/L
Median 153 183 0.744
Interquartile range 96–241 45–292
Platelets ⫻ 109/L
Median 500 476 0.59
Interquartile range 370–640 352–682
Hb, g/dL
Median 10.1 10.05 0.740
Interquartile range 8.7–11.4 8.6–11.2
Pleural fluid LDH, U/L n ⫽ 17 n ⫽ 24
Median 10,000 6,953 0.368
Interquartile range 4880–20,000 2,992–16,554
Pleural fluid albumin, g/L n ⫽ 17 n ⫽ 24
Median 21 21.5 0.151
Interquartile range 20–29 20–46
Pleural fluid WBC n ⫽ 24 n ⫽ 23
⫹/⫺ 21/3 21/2 0.79
Fever days after intervention 2.5 (0–10) 2.5 (0–25) 0.635
O2 supplementation 12 12 1
Microbiology (positive blood or fluid
culture and/or PCR)
S. pneumoniae 18 13 0.196
Other organisms 3 6 0.317

Definition of abbreviations: CRP ⫽ C-reactive protein; LDH, lactate dehydrogenase; PCR ⫽ polymerase chain reaction; VATS ⫽
video-assisted thoracoscopic surgery; WBC ⫽ white blood cells.

to be useful in evaluating the success of treatment. However,
our study was not designed to evaluate this outcome and there-
fore may have lacked the power to detect a difference between
the groups. A study of adult patients also found that ultrasound
was unable to accurately identify the stage of the disease, using
Light’s criteria as the gold standard (49). Light’s criteria (50)
have not been fully evaluated in children, and it is difficult to
know whether they are applicable to the pediatric population.
There was no statistically significant difference in all the second-
ary outcome measures except costs between the two groups.
Although there was a borderline statistical difference in the
number of chest drain in situ days between the two groups in
favor of the VATS group, no significant clinical difference was
demonstrated, as this did not have an impact on the hospital stay
after intervention. The protocol is biased against the urokinase
group, because intrapleural urokinase is administered over 3 d.
The failure rates were similar in both the treatment arms;
that is, five patients in the urokinase arm did not respond to
conservative management and had to proceed to have secondary
VATS with or without mini-thoracotomy, and four patients in
the VATS arm had the operation converted to a mini-thoracotomy.
Interestingly, four of the five urokinase failures were from “Win-
ter 2003,” which may reflect a change in virulence of pathogens
from winter to winter. These results (ⵑ 16% failure rate) reflect
the failure rates reported in various other studies: 6–15% for
chest drains with fibrinolytics (11, 15, 40) and 0–20% for primary
VATS (16, 20, 32, 51).
The commonest causative organism isolated in our study was
fully penicillin-sensitive S. pneumoniae, suggesting that the in-
crease in the incidence of empyema is not due to the emergence
of penicillin-resistant strains. There was no impact of the caus-
ative organism on the length of hospital stay.
We acknowledge that there are some limitations to this study.
The study reflects single-center practice; however, patients are
diverse and representative of the United Kingdom as a whole,
as referrals are from the south-east of England and VATS was
performed by three different surgeons trained in three different
institutions.
Radiologic follow-up at 6 mo was somewhat arbitrary, as
no validated CXR scoring systems for evaluation of radiologic
resolution of empyema exist. However, the CXR changes that
were evaluated in our study were the common changes looked
for on radiographic assessment after infection, and there was
considerable agreement between our two radiologists. The com-
monest abnormality noted was pleural shadowing, which was
seen in 38 (86%) of the 44 CXRs assessed.
No functional outcome measures were included in the data
analysis, although spirometry and ventilation perfusion (V̇/Q̇)
scan at 6 mo after intervention were included in the study design,
with ethical approval. Significant proportions (71%) of the re-
cruited patients were under 6 yr of age, and it would have been
difficult to measure lung function by standard spirometry. We
were unable to perform V̇/Q̇ scans in a significant number of
the patients due to parental reluctance at subjecting a well child
226 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 174 2006
at 6 mo after intervention to a perceived invasive procedure
(intravenous cannulation for injecting radioisotope for V̇/Q̇
scan).
An important observation is that VATS is $2,250, or 25%
more expensive than intrapleural urokinase at our center without
including additional start up costs for provision of VATS service.
The main limitations of this analysis are that it only considers
costs and not effectiveness; each cost estimate is based on only
30 patients; and the prices used are from a single institution in
the United Kingdom. However, it is unlikely that there is a
significant cost differential between the two treatments in other
developed countries. All patients undergoing VATS were as-
sumed to have a CT chest scan, as most surgeons request a
preoperative CT chest scan and it is not our practice to routinely
request CT chest scans for treatment with intrapleural urokinase.
Even if costs of CT chest scans were omitted from the analysis,
VATS was still significantly more expensive than intrapleural
urokinase. The cost analysis does strongly suggest that VATS
will only be less expensive if the probability of treatment success
is much lower with urokinase than with VATS. Our study has
shown that both intrapleural urokinase and primary VATS are
equally effective for the treatment of childhood empyema, and
therefore VATS is unlikely to be less expensive.
There have been no reports addressing the financial burden
to health services, should VATS be the favored approach for the
treatment of empyema in children. The incidence of childhood
pneumonia is in the order of 10–15 cases per 1,000 children (52).
There were an estimated 12 million children aged 0–16 yr in the
United Kingdom (www.statistics.gov.uk) and 74 million children
in the United States (www.childstats.gov) in mid-2005. If 0.6%
of the pneumonias progressed to empyemas, this would mean
1,080 and 6,660 cases of childhood empyema per year in the
United Kingdom and the United States, respectively. This
broadly would indicate a saving of $2.5 million and $15 million
per annum to the UK and U.S. health services, respectively, if
intrapleural urokinase is the first line of treatment for childhood
empyema compared with VATS.
In conclusion, we have provided an evidence base to guide
the management of childhood empyema. Intrapleural urokinase
should be the primary treatment of choice in the treatment of
uncomplicated empyemas in children. This will have further
implications on reducing the financial burden of costs to the
health services.
Conflict of Interest Statement : None of the authors has a financial relationship
with a commercial entity that has an interest in the subject of this manuscript.
Acknowledgment : The authors thank Dr. Robert Dinwiddie, Dr. Colin Wallis,
Dr. Paul Aurora (Respiratory Pediatricians); Dr. John Hartley, Dr. Kathryn Harris
(Microbiologists); Dr. Alistair Calder (Radiologist); Ms. Deborah Ridout (Statistical
Support); and Ms. Moranti Falade (Service Agreement and Costing).
References
1. Rees JH, Spencer DA, Parikh D, Weller P. Increase in incidence of
childhood empyema in West Midlands, UK. Lancet 1997;349:402.
2. Byington CL, Spencer LY, Johnson TA, Pavia AT, Allen D, Mason EO,
Kaplan S, Carroll KC, Daly JA, Christenson JC, et al. An epidemiologi-
cal investigation of a sustained high rate of pediatric parapneumonic
empyema: risk factors and microbiological associations. Clin Infect
Dis 2002;34:434–440.
3. Balfour-Lynn IM, Abrahamson E, Cohen G, Hartley J, King S, Parikh
D, Spencer D, Thomson AH, Urquhart D. BTS guidelines for the
management of pleural infection in children. Thorax 2005;60:i1–21.
4. Hardie W, Bokulic R, Garcia VF, Reising SF, Christie CD. Pneumococcal
pleural empyemas in children. Clin Infect Dis 1996;22:1057–1063.
5. Balfour-Lynn IM. Some consensus but little evidence: guidelines on man-
agement of pleural infection in children. Thorax 2005;60:94–96.
6. Krishnan S, Amin N, Dozor AJ, Stringel G. Urokinase in the management
of complicated parapneumonic effusions in children. Chest 1997;112:
1579–1583.
7. Kornecki A, Sivan Y. Treatment of loculated pleural effusion with in-
trapleural urokinase in children. J Pediatr Surg 1997;32:1473–1475.
8. Moulton JS, Benkert RE, Weisiger KH, Chambers JA. Treatment of
complicated pleural fluid collections with image-guided drainage and
intracavitary urokinase. Chest 1995;108:1252–1259.
9. Chan PW, Crawford O, Wallis C, Dinwiddie R. Treatment of pleural
empyema. J Paediatr Child Health 2000;36:375–377.
10. Maskell NA, Davies CW, Nunn AJ, Hedley EL, Gleeson FV, Miller R,
Gabe R, Rees GL, Peto TE, Woodhead MA, et al. UK Controlled
trial of intrapleural streptokinase for pleural infection. N Engl J Med
2005;352:865–874.
11. Thomson AH, Hull J, Kumar MR, Wallis C, Balfour L. I. Randomised
trial of intrapleural urokinase in the treatment of childhood empyema.
Thorax 2002;57:343–347.
12. Singh M, Mathew JL, Chandra S, Katariya S, Kumar L. Randomized
controlled trial of intrapleural streptokinase in empyema thoracis in
children. Acta Paediatr 2004;93:1443–1445.
13. Jaffe A, Cohen G. Thoracic empyema. Arch Dis Child 2003;88:839–841.
14. Kern JA, Rodgers BM. Thoracoscopy in the management of empyema
in children. J Pediatr Surg 1993;28:1128–1132.
15. Doski JJ, Lou D, Hicks BA, Megison SM, Sanchez P, Contidor M,
Guzzetta PC Jr. Management of parapneumonic collections in infants
and children. J Pediatr Surg 2000;35:265–268.
16. Tonz M, Ris HB, Casaulta C, Kaiser G. Is there a place for thoracoscopic
debridement in the treatment of empyema in children? Eur J Pediatr
Surg 2000;10:88–91.
17. Davidoff AM, Hebra A, Kerr J, Stafford PW. Thoracoscopic manage-
ment of empyema in children. J Laparoendosc Surg 1996;6:S51–S54.
18. Rodriguez JA, Hill CB, Loe WA Jr, Kirsch DS, Liu DC. Video-assisted
thoracoscopic surgery for children with stage II empyema. Am Surg
2000;66:569–572.
19. Merry CM, Bufo AJ, Shah RS, Schropp KP, Lobe TE. Early definitive
intervention by thoracoscopy in pediatric empyema. J Pediatr Surg
1999;34:178–180.
20. Subramaniam R, Joseph VT, Tan GM, Goh A, Chay OM. Experience
with video-assisted thoracoscopic surgery in the management of com-
plicated pneumonia in children. J Pediatr Surg 2001;36:316–319.
21. Meier AH, Smith B, Raghavan A, Moss RL, Harrison M, Skarsgard E.
Rational treatment of empyema in children. Arch Surg 2000;135:907–
912.
22. Kercher KW, Attorri RJ, Hoover JD, Morton D Jr. Thoracoscopic decor-
tication as first-line therapy for pediatric parapneumonic empyema: a
case series. Chest 2000;118:24–27.
23. Patton RM, Abrams RS, Gauderer MW. Is thoracoscopically aided pleu-
ral debridement advantageous in children? Am Surg 1999;65:69–72.
24. Grewal H, Jackson RJ, Wagner CW, Smith SD. Early video-assisted
thoracic surgery in the management of empyema. Pediatrics 1999;103:
e63.
25. Wait MA, Sharma S, Hohn J, Dal Nogare A. A randomized trial of
empyema therapy. Chest 1997;111:1548–1551.
26. Ferguson AD, Prescott RJ, Selkon JB, Watson D, Swinburn CR. The
clinical course and management of thoracic empyema. QJM 1996;89:
285–289.
27. Sonnappa S, Cohen G, van Doorn C, Elliott M, Owens C, Jaffe A.
Randomised trial comparing chest drain with intrapleural urokinase
versus video assisted thoracoscopic surgery for the treatment of empy-
ema in children. Thorax 2005;60:II30–II31.
28. Carey JA, Hamilton JR, Spencer DA, Gould K, Hasan A. Empyema
thoracis: a role for open thoracotomy and decortication. Arch Dis
Child 1998;79:510–513.
29. King S, Thomson A. Radiological perspectives in empyema. Br Med Bull
2002;61:203–214.
30. Saglani S, Harris KA, Wallis C, Hartley JC. Empyema: the use of broad
range 16S rDNA PCR for pathogen detection. Arch Dis Child 2005;
90:70–73.
31. Barbato A, Panizzolo C, Monciotti C, Marcucci F, Stefanutti G, Gamba
PG. Use of urokinase in childhood pleural empyema. Pediatr Pulmonol
2003;35:50–55.
32. Kilic N, Celebi S, Gurpinar A, Hacimustafaoglu M, Konca Y, Ildirim I,
Dogruyol H. Management of thoracic empyema in children. Pediatr
Surg Int 2002;18:21–23.
33. Rosen H, Nadkarni V, Theroux M, Padman R, Klein J. Intrapleural
streptokinase as adjunctive treatment for persistent empyema in pedi-
atric patients. Chest 1993;103:1190–1193.
34. Stringel G, Hartman AR. Intrapleural instillation of urokinase in the
treatment of loculated pleural effusions in children. J Pediatr Surg
1994;29:1539–1540.
Sonnappa, Cohen, Owens, et al.: Urokinase and VATS in Childhood Empyema
35. Wells RG, Havens PL. Intrapleural fibrinolysis for parapneumonic effu-
sion and empyema in children. Radiology 2003;228:370–378.
36. Cochran JB, Tecklenburg FW, Turner RB. Intrapleural instillation of
fibrinolytic agents for treatment of pleural empyema. Pediatr Crit Care
Med 2003;4:39–43.
37. Hawkins JA, Scaife ES, Hillman ND, Feola GP. Current treatment of
pediatric empyema. Semin Thorac Cardiovasc Surg 2004;16:196–200.
38. Gates RL, Hogan M, Weinstein S, Arca MJ. Drainage, fibrinolytics, or
surgery: a comparison of treatment options in pediatric empyema.
J Pediatr Surg 2004;39:1638–1642.
39. Balci AE, Eren S, Ulku R, Eren MN. Management of multiloculated
empyema thoracis in children: thoracotomy versus fibrinolytic treat-
ment. Eur J Cardiothorac Surg 2002;22:595–598.
40. Ulku R, Onat S, Kilic N. Intrapleural fibrinolytic treatment of multilocu-
lated pediatric empyemas. Minerva Pediatr 2004;56:419–423.
41. Ozcelik C, Inci I, Nizam O, Onat S. Intrapleural fibrinolytic treatment
of multiloculated postpneumonic pediatric empyemas. Ann Thorac
Surg 2003;76:1849–1853.
42. Barnes NP, Hull J, Thomson AH. Medical management of parapneu-
monic pleural disease. Pediatr Pulmonol 2005;39:127–134.
43. Bouros D, Antoniou KM, Chalkiadakis G, Drositis J, Petrakis I, Siafakas
N. The role of video-assisted thoracoscopic surgery in the treatment
227
of parapneumonic empyema after the failure of fibrinolytics. Surg
Endosc 2002;16:151–154.
44. Sit SC, Cohen G, Jaffe A. Urokinase in the treatment of childhood
empyema. Thorax 2003;58:93–94.
45. Hull J, Thomson A. Empyema thoracis: a role for open thoracotomy
and decortication. Arch Dis Child 1999;80:581.
46. Yim AP, Wan S, Lee TW, Arifi AA. VATS lobectomy reduces cytokine
responses compared with conventional surgery. Ann Thorac Surg
2000;70:243–247.
47. Sedrakyan A, van der Meulen J, Lewsey J, Treasure T. Variation in use
of video assisted thoracic surgery in the UK. BMJ 2004;329:1011–1012.
48. McCulloch P. Half full or half empty VATS? BMJ 2004;329:1012.
49. Kearney SE, Davies CW, Davies RJ, Gleeson FV. Computed tomography
and ultrasound in parapneumonic effusions and empyema. Clin Radiol
2000;55:542–547.
50. Light RW, Macgregor MI, Luchsinger PC, Ball WC Jr. Pleural effusions:
the diagnostic separation of transudates and exudates. Ann Intern Med
1972;77:507–513.
51. Cohen G, Hjortdal V, Ricci M, Jaffe A, Wallis C, Dinwiddie R, Elliott
MJ, de Leval MR. Primary thoracoscopic treatment of empyema in
children. J Thorac Cardiovasc Surg 2003;125:79–83.
52. Farha T, Thomson AH. The burden of pneumonia in children in the
developed world. Paediatr Respir Rev 2005;6:76–82.