- Email: [email protected]
‘Less may be best’—Pediatric parapneumonic effusion and empyema management: Lessons from a UK center
Journal of Pediatric Surgery xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg
‘Less may be best’—Pediatric parapneumonic effusion and empyema management: Lessons from a UK center Anna-May Long a, Jonathan Smith-Williams b, Sarah Mayell c, Jon Couriel c, Matthew O. Jones a, Paul D. Losty a,d,⁎ a
Department of Paediatric Surgery, Alder Hey Children's Hospital NHS Foundation Trust, Liverpool, UK Medical School University of Liverpool, Liverpool, UK Department of Respiratory Medicine, Alder Hey Children's Hospital NHS Foundation Trust, Liverpool, UK d Academic Department of Paediatric Surgery, University of Liverpool, Liverpool, UK b c
a r t i c l e
i n f o
Article history: Received 12 May 2015 Received in revised form 31 July 2015 Accepted 31 July 2015 Available online xxxx Key words: Empyema Parapneumonic effusion Urokinase Antibiotics Thoracic surgery Clinical trials
a b s t r a c t Background: Children with empyema are managed at our center using a protocol-driven clinical care pathway. Chemical fibrinolysis is deployed as first-line management for significant pleural disease. We therefore examined clinical outcome(s) to benchmark standards of care while analyzing disease severity with introduction of the pneumococcal conjugate vaccine. Methods: Medical case-records of children managed at a UK pediatric center were surveyed from Jan 2006 to Dec 2012. Binary logistic regression was utilized to study failure of fibrinolytic therapy. The effects of age, comorbidity, number of days of intravenous antibiotics prior to drainage and whether initial imaging showed evidence of necrotizing disease were also studied. Results: A total of 239 children were treated [age range 4 months–19 years; median 4 years]. A decreasing number of patients presenting year-on-year since 2006 with complicated pleural infections was observed. The majority of children were successfully managed without surgery using antibiotics alone (27%) or a fine-bore chest-drain and urokinase (71%). Only 2% of cases required primary thoracotomy. 14.7% cases failed fibrinolysis and required a second intervention. The only factor predictive of failure and need for surgery was suspicion of necrotizing disease on initial imaging (P = 0.002, OR 8.69). Conclusion: Pediatric patients with pleural empyema have good outcomes when clinical care is led by a multidisciplinary team and protocol driven care pathway. Using a ‘less is best’ approach few children require surgery. © 2015 Elsevier Inc. All rights reserved.
Pleural empyema is a lung disease occurring in children where parenchymal infection is complicated by parapneumonic effusion. Exudative fluid accumulates in the pleural cavity and deposition of fibrin occurs. Over time pleural surfaces are coated with an inelastic ‘peel’ inhibiting lung expansion [1]. Where this is causing significant physiological compromise to patients expedient drainage of the effusion and debridement of fibrin peel using either mechanical or chemical methods may be required [2,3]. Chemical fibrinolysis is encouraged as first line therapy for complicated disease [2,3]. Randomized controlled trials show equivalent outcome(s) comparing (I) decortication by thoracoscopy (VATS) versus (II) chemical fibrinolyis—with significant cost savings with the latter [4,5]. To this end, guidelines have been developed by the British Thoracic Society (BTS) and the American Pediatric Surgical Association (APSA). This study analyzes and reports patient outcomes from a major UK pediatric center against the background of developing a protocol driven integrated clinical care pathway for empyema. We have examined the ⁎ Corresponding author at: Alder Hey Children's Hospital NHS Foundation Trust, University of Liverpool, Liverpool, UK. Tel.: +44 151 228 4811. E-mail address: [email protected] (P.D. Losty).
progress of all children treated at our center focusing on (I) length of hospital stay, (II) readmission rate(s), and (III) need for second intervention(s). Factors potentially related to outcome such as (I) age, (II) comorbidity, and (III) infecting organism(s) were examined in detail. Any feature(s) relating to failure of chemical fibrinolysis was also scrutinized with the aim of identifying patient subgroups unsuitable for fibrinolysis as primary therapy. In September 2006, the seven valent pneumococcal conjugate vaccine (PCV7) was added to routine childhood immunization protocols in the UK [6]. In March 2010 the vaccine was replaced by a thirteen valent pneumococcal conjugate vaccine after public health concerns were raised regarding a growing predominance of invasive serotypes not covered by PCV7 [7]. This study therefore also reports microbiology findings and outcome metrics in the postvaccination era. At our center after a diagnosis of empyema is confirmed by ultrasound all children are then managed by a multidisciplinary (MDT) team led by pediatric respiratory physicians and surgeons. CT chest scans are not a routinely deployed investigation in ‘work up’. Here CT is reserved only for patients where a bronchopleural fistula is suspected or an underlying congenital lung lesion, eg, congenital cystic adenomatoid malformation (CCAM). CT is further used in patient care
http://dx.doi.org/10.1016/j.jpedsurg.2015.07.022 0022-3468/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Long A-M, et al, ‘Less may be best’—Pediatric parapneumonic effusion and empyema management: Lessons from a UK center, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.07.022
2
A-M. Long et al. / Journal of Pediatric Surgery xxx (2015) xxx–xxx
children b10 kg and 40,000 IU diluted in 40 ml of 09.9% saline for children N 10 kg). Primary thoracotomy is reserved for children with a suspicion of bronchopleural fistula(e) and severe necrotizing lung disease with severe systemic illness or in cases with a suspected lung lesion.
Table 1 Population characteristics. Characteristic Sex: Male Female Median age (IQR) Age range Severe comorbidity No Yes ICU/HDU Admission No Yes Causative organism S. pneumoniae Not identified Group A Strep. S. aureus Othersa
n (%) 122 (51) 117 (49) 4 years, (2–8 years) 4 months–19 years 209 (87) 30 (13) 187 (78) 52 (21) 112 (56) 61 (26) 19 (8) 4 (2) 5 (2)
Primary management strategy IV antibiotics alone Chest drain and urokinase Primary thoracotomy Total a b
Failed primary treatment (required second procedure) 66 (27) 169 (71) 4 (2) 239
2b (3) 25 (14.7) 0 27 (11.3)
Mycobacterium tuberculosis 1, Enterococcus faecium 1, Streptococcus intermedius 3. Readmitted and required thoracotomy.
plans when a primary or rescue thoracotomy is considered. Clinically stable patients with small pleural collections seen on ultrasound imaging are therefore managed with antibiotics and active observation. Cefuroxime is employed as a first-line antibiotic and oral clarithromycin added in patients N5 years. Blood cultures, serology, Anti-streptolysin O titre (ASOT), and blood pneumococcal PCR are obtained on all patients at first presentation with pleural fluid routinely sent to the laboratory(s) for microscopy, culture and sensitivity including pneumococcal PCR. Patients with persistently high fever and significant pleural collections are then managed with an 8 Fr chest drain inserted under general anesthesia using the Seldinger technique with instillation of intrapleural urokinase. Six doses of urokinase are administered at 12 hourly intervals (10,000 IU diluted in 10 ml of 0.9% saline was used for
Methods This clinical study was Institutional Review Board (IRB) approved by Alder Hey Childrens Hospital (in accordance with the Declaration of Helsinki 1964). Medical case records were analyzed on patients admitted to our center with a diagnosis of empyema during the era(s) January 2006–December 2012. All patients were managed using the protocol driven MDT care pathway. Data examined were (I) demographics (II), patient comorbidity (III) infecting organism[s], (IV) radiology, and (V) antibiotic therapy. Length of hospital stay, admission to the intensive care unit (ICU) or high dependency (HDU), hospital readmission and need for second intervention(s) were examined. Data on readmission and postprocedural length of stay were fully available from 2007 onward. The influence of severe patient comorbidity(s) and microorganism(s) on hospital stay and admission to ICU or HDU was analyzed with Mann– Whitney U, Chi χ2 and Fishers Exact tests. Factors linked to fibrinolytic failure were identified prospectively: (I) patient age, (II) significant comorbidity, (III) duration of intravenous antibiotic[s] prior to drainage, and (IV) suspicion of necrotizing disease on initial chest imaging. Covariates were selected a priori based on clinical hypotheses. Significant comorbidity included (a) chromosomal anomaly, (b) neurological disability, (c) congenital heart disease, (d) connective tissue disorders, and (e) immune deficiency states. Necrotizing disease on initial imaging was defined as evidence of parenchymal destruction, suppuration or air in the pleural space with chest x-ray, ultrasound or CT. The influence of these factors was further assessed with binary logistic regression to undertake univariate analysis. All statistical analyses were performed using SPSS 20 (with P b 0.05 equals significant). Results A total of 239 patients were admitted to our hospital with empyema in the era(s) 2006–2012. Table 1 shows the demographics with primary management strategies for the cohort (age range 4 months–19 years: median 4 years). The proportion of those who had significant
Fig. 1. Bar chart illustrating empyema hospital admissions per year and management.
Please cite this article as: Long A-M, et al, ‘Less may be best’—Pediatric parapneumonic effusion and empyema management: Lessons from a UK center, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.07.022
A-M. Long et al. / Journal of Pediatric Surgery xxx (2015) xxx–xxx Table 2 Univariate analysis of factors related to failure of fibrinolytic therapy. Factor
Odds ratio
(95% CI)
P value
Age Presence of comorbidity Prior duration of IV antibiotics Suspicion of necrotizing disease on initial imaging
0.96 1.7 0.98 8.69
0.87–1.1 0.54–5.2 0.85–1.12 2.15–35.0
0.51 0.31 0.77 0.002*
comorbidity(s) was 12.6%. First-line therapy involved antibiotics alone in 27% cases; pigtail chest drain and urokinase in 71% with only 2% requiring surgery. Here a primary thoracotomy was undertaken in 4 cases. One patient had an underlying congenital lung lesion that became infected and 3 operations were required emergently for lung abscess/ necrotizing disease with bronchopleural fistulae. The causative microorganisms were identified in 140 patients (59%; Table 1). These included Streptococcus pneumoniae (56%), Group A streptococcus (8%), and Staphylococcus aureus (2%) with a single patient here having MRSA. A declining number of empyema admissions year-on-year was observed with no obvious change in disease severity at first presentation. We did not encounter an increase in the number of cases requiring surgery, ie, early thoracotomy (Fig. 1). Median hospital stay was 10 days [IQR 8–13 days]—with a postprocedural length of stay in those cases requiring intervention of 9 days [IQR 7–12 days]. Patients with significant comorbidity(s) had a significantly longer hospital stay [median 14 days vs median 10 days in otherwise healthy children (P = 0.001)]. Children with comorbidity(s) were significantly more likely also to require ICU or HDU admission—47% vs 18% of otherwise healthy children (P = 0.001). It was also noted that those with pleural disease linked with Group A streptococcus stayed significantly longer in hospital vs cases attributed to S. pneumoniae infection [median 15 days vs 10 days—P b 0.001]. Patients with Group A streptococcal disease also had more severe illness with 58% requiring HDU or ICU admissions compared with 19% cases infected by S. pneumoniae (P = 0.001). Six percent of children were readmitted to hospital (data based on complete analyses from 178 cases). Median time to readmission was 6 days. Of the 11 cases requiring hospital readmission, 5 required thoracotomy—2 of these children had previously been managed with antibiotics alone and 3 with a chest drain and fibrinolysis only. The overall failure rate of the primary treatment modality was 11.3%. Of those who were treated with fibrinolytic therapy as first-line therapy, 25 cases (14.7%) required a second procedure. Twenty patients had a thoracotomy, 2 had thoracoscopic decortication (VATS) and 3 required insertion of a second chest drain We further analyzed the cohort that were treated initially with a chest drain and urokinase and required a second
3
intervention. Factors influencing failure of primary management were assessed with univariate analysis using binary logistic regression (Table 2). The only factor found to be statistically significant and associated with the need for second intervention after fibrinolysis failure was the existence of necrotizing disease suspected from primary radiology imaging (P = 0.002, OR 8.69). There were 14 patients with such findings in the entire study—2 cases were treated initially with primary thoracotomy, 3 with antibiotics alone and 9 patients with a chest drain and urokinase. Of those treated with a chest drain and urokinase, 5 required a second procedure. Three underwent thoracotomy and decortication, a single patient had thoracoscopic debridement (VATS) and insertion of a larger drain and 1 case had a second chest drain inserted. The 5 cases who required a second intervention had a median length of stay of 15 days. In the study one patient died giving an overall mortality rate from thoracic empyema of 0.4% at this center. The single fatality had significant comorbidity—severe neurological disorder with chronic existing lung disease and developed progressive respiratory failure 2 months after treatment of empyema. Discussion This study has highlighted the contemporary outcome(s) of a large cohort of 239 children with pleural empyema managed at a UK pediatric center. All patients were triaged in a consistent way using a MDT protocol-driven care plan. The care pathway illustrates that a ‘less may be best’ policy works for the vast majority of children with early deployment of a fine bore chest drain and fibrinolysis as first line treatment [8]. Outcomes herein reported included factors influencing hospital stay and need for readmission and indication(s) for secondary intervention. The majority of cases were all successfully managed with a chest drain and fibrinolysis with only 14.7% of the cohort requiring a second procedure. These findings compare favorably to several published studies [4,5]. Median postprocedural length of stay was 9 days with a hospital readmission rate of 6.2%. Table 3 provides a summary of outcomes from several published studies including the present Liverpool series. Children with significant comorbidity(s) required an additional median hospital stay of 4 days longer than otherwise previously healthy patients. These cases were also significantly more likely to require ICU or HDU admission. We also found that the infecting microrganism had a significant influence on the likelihood of being admitted to ICU or HDU. Herein the majority of patients presenting with Group A streptococcal infection required ICU admission reflecting a causal association of invasive streptococcal disease with severe systemic toxin mediated illness—‘toxic shock syndrome’ [9]. We observed a falling number of patients admitted to our hospital year-on-year after introduction of the pneumococcal conjugate vaccine with no apparent increase in complexity of disease. This phenomenon
Table 3 Outcome measures from other published studies utilising chemical fibrinolysis. Author (year)
Study type
Protocol
LOS (d)
T/PP
Failure rate (%)
Thomson et al. (2002) [14] Gates et al. (2004) [15] Gates et al. (2004) [16]
RCT SR RR
Drain (both fine-bore and large) + UK Chest Drain + UK or SK Chest Drain and TpA
RCT RCT Database analysis RR RR RR
Drain + UK Drain + TpA Chest drain + fibrinolysis Chest drain + TpA 8 Fr Pigtail + UK 8 Fr Pigtail + UK
PP PP T PP PP PP – PP PP T PP
6.66 N/A N/A
Sonnappa et al. (2006) [4] St Peter et al. (2009) [5] Goldin et al. (2012) [17] Gasior et al. (2013) [18] Petel et al. (2013) [19] Long et al. (2015) (this study)
7.4a 18.9b 8b 6.2b 6b 6.8b N/A 7.2a 11.1b 10b 9b
16.6 16.6 21 15.7 40 14.7
SR, systematic review; RR, retrospective review; RCT, randomized controlled trial; LOS, length of stay; T, total length of stay; PP, postprocedural length of stay; UK, urokinase; SK, streptokinase; TpA, tissue plasminogen activating factor; N/A, not assessed. a Mean. b Median.
Please cite this article as: Long A-M, et al, ‘Less may be best’—Pediatric parapneumonic effusion and empyema management: Lessons from a UK center, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.07.022
4
A-M. Long et al. / Journal of Pediatric Surgery xxx (2015) xxx–xxx
has been noted in other European countries perhaps suggesting some herd protection in those N2 years old who are not routinely offered vaccination [10,11]. S. pneumoniae however remained the most common infecting organism accounting for 56% of the index cases reported in this study. When we analyzed and sought to identify which children were not suitable for chest drainage and urokinase as a first-line treatment for empyema it was only the suspicion of necrotizing disease on initial chest imaging that was associated with an increased treatment failure rate (55% vs 12.4% without). Even so, 45% of this subgroup with the suspicion of necrotizing disease were mostly managed with fibrinolytics and did not require further interventions. To this end, we have not changed our practice to offer such patients who are more severely ill ‘up front’ primary thoracotomy. The information derived from this subgroup analysis is however an important factor to consider particularly when counselling families as the chance of therapy success (%) with chemical fibrinolysis alone is considerably lower versus those children without suspicion of necrotizing disease. We have previously published several report(s) highlighting experience with pediatric empyema some years ago. Evolving practice over twenty years with an MDT team (physicians and surgeons) and the ready availability of a protocol driven care pathway strongly convinces us that ‘less may be best’ for pleural empyema. It is therefore now possible for many children to avoid major surgical operation(s) notably VATS or thoracotomy for empyema [4,5,8,12,13].
References [1] Hamm H, Light RW. Parapneumonic effusion and empyema. Eur Respir J 1997;10: 1150–6. [2] Balfour-Lynn IM, Abrahamson E, Cohen G, et al. BTS guidelines for the management of pleural infection in children. Thorax 2005;60(Suppl. 1):i1–i21.
[3] Islam S, Calkins CM, Goldin AB, et al. APSA outcomes and clinical trials committee. The diagnosis and management of empyema in children: a comprehensive review from the APSA Outcomes and Clinical Trials Committee. J Pediatr Surg 2012;47: 2101–10. [4] Sonnappa S, Cohen G, Owens CM, et al. Comparison of urokinase and video-assisted thoracoscopic surgery for treatment of childhood empyema. Am J Respir Crit Care Med 2006;174:221–7. [5] St Peter SD, Tsao K, Spilde TL, et al. Thoracoscopic decortication versus tube thoracostomy with fibrinolysis for empyema in children: a prospective, randomized trial. J Pediatr Surg 2009;44:106–11. [6] Public Health England. Immunsation against infectious diseases. https://www.gov. uk/government/collections/immunisation-against-infectious-diseasethe-greenbook. [Accessed May 2nd 2014]. [7] Hanquet G, Kissling E, Fenoll A, et al. Pneumococcal serotypes in children in 4 European countries. Emerg Infect Dis 2010;16:1428–39. [8] Khalil BA, Corbett PA, Jones MO, et al. Less is best? The impact of urokinase as the first line management of empyema thoracis. Pediatr Surg Int 2007;23:129–33. [9] Lappin E, Ferguson AJ. Gram-positive toxic shock syndromes. Lancet Infect Dis 2009; 9:281–90. [10] Angoulvant F, Pereira M, Perreaux F, et al. Early impact of 13-valent pneumococcal conjugate vaccine on community-acquired pneumonia in children. Clin Infect Dis 2014;58:918–24. [11] Klugman KP. Editorial commentary: a tale of 2 pneumococcal vaccines. Clin Infect Dis 2014;58(7):925–7. [12] Shankar KR, Kenny SE, Okoye BO, et al. Evolving experience in the management of empyema thoracis. Acta Paediatr 2000;89:417–20. [13] Proesmans M, Gijens B, Van de Wijdeven P, et al. Clinical outcome of parapneumonic empyema in children treated according to a standardized medical treatment. Eur J Pediatr 2014;173:1339–45. [14] Thomson AH, Hull J, Kumar MR, et al. Randomised trial of intrapleural urokinase in the treatment of childhood empyema. Thorax 2002;57:343–7. [15] Gates RL, Caniano DA, Hayes JR, et al. Does VATS provide optimal treatment of empyema in children? A systematic review. J Pediatr Surg 2004;39:381–6. [16] Gates RL, Hogan M, Weinstein S, et al. Drainage, fibrinolytics, or surgery: a comparison of treatment options in pediatric empyema. J Pediatr Surg 2004;39:1638–42. [17] Goldin AB, Parimi C, LaRiviere C, et al. Outcomes associated with type of intervention and timing in complex pediatric empyema. Am J Surg 2012;203:665–7. [18] Gasior AC, Knott EM, Sharp SW, et al. Experience with an evidence-based protocol using fibrinolysis as first line treatment for empyema in children. J Pediatr Surg 2013;48:1312–5. [19] Petel D, Li P, Emil S. Percutaneous pigtail catheter versus tube thoracostomy for pediatric empyema: a comparison of outcomes. Surgery 2013;154:655–60.
Please cite this article as: Long A-M, et al, ‘Less may be best’—Pediatric parapneumonic effusion and empyema management: Lessons from a UK center, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.07.022
Contents lists available at ScienceDirect
Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg
‘Less may be best’—Pediatric parapneumonic effusion and empyema management: Lessons from a UK center Anna-May Long a, Jonathan Smith-Williams b, Sarah Mayell c, Jon Couriel c, Matthew O. Jones a, Paul D. Losty a,d,⁎ a
Department of Paediatric Surgery, Alder Hey Children's Hospital NHS Foundation Trust, Liverpool, UK Medical School University of Liverpool, Liverpool, UK Department of Respiratory Medicine, Alder Hey Children's Hospital NHS Foundation Trust, Liverpool, UK d Academic Department of Paediatric Surgery, University of Liverpool, Liverpool, UK b c
a r t i c l e
i n f o
Article history: Received 12 May 2015 Received in revised form 31 July 2015 Accepted 31 July 2015 Available online xxxx Key words: Empyema Parapneumonic effusion Urokinase Antibiotics Thoracic surgery Clinical trials
a b s t r a c t Background: Children with empyema are managed at our center using a protocol-driven clinical care pathway. Chemical fibrinolysis is deployed as first-line management for significant pleural disease. We therefore examined clinical outcome(s) to benchmark standards of care while analyzing disease severity with introduction of the pneumococcal conjugate vaccine. Methods: Medical case-records of children managed at a UK pediatric center were surveyed from Jan 2006 to Dec 2012. Binary logistic regression was utilized to study failure of fibrinolytic therapy. The effects of age, comorbidity, number of days of intravenous antibiotics prior to drainage and whether initial imaging showed evidence of necrotizing disease were also studied. Results: A total of 239 children were treated [age range 4 months–19 years; median 4 years]. A decreasing number of patients presenting year-on-year since 2006 with complicated pleural infections was observed. The majority of children were successfully managed without surgery using antibiotics alone (27%) or a fine-bore chest-drain and urokinase (71%). Only 2% of cases required primary thoracotomy. 14.7% cases failed fibrinolysis and required a second intervention. The only factor predictive of failure and need for surgery was suspicion of necrotizing disease on initial imaging (P = 0.002, OR 8.69). Conclusion: Pediatric patients with pleural empyema have good outcomes when clinical care is led by a multidisciplinary team and protocol driven care pathway. Using a ‘less is best’ approach few children require surgery. © 2015 Elsevier Inc. All rights reserved.
Pleural empyema is a lung disease occurring in children where parenchymal infection is complicated by parapneumonic effusion. Exudative fluid accumulates in the pleural cavity and deposition of fibrin occurs. Over time pleural surfaces are coated with an inelastic ‘peel’ inhibiting lung expansion [1]. Where this is causing significant physiological compromise to patients expedient drainage of the effusion and debridement of fibrin peel using either mechanical or chemical methods may be required [2,3]. Chemical fibrinolysis is encouraged as first line therapy for complicated disease [2,3]. Randomized controlled trials show equivalent outcome(s) comparing (I) decortication by thoracoscopy (VATS) versus (II) chemical fibrinolyis—with significant cost savings with the latter [4,5]. To this end, guidelines have been developed by the British Thoracic Society (BTS) and the American Pediatric Surgical Association (APSA). This study analyzes and reports patient outcomes from a major UK pediatric center against the background of developing a protocol driven integrated clinical care pathway for empyema. We have examined the ⁎ Corresponding author at: Alder Hey Children's Hospital NHS Foundation Trust, University of Liverpool, Liverpool, UK. Tel.: +44 151 228 4811. E-mail address: [email protected] (P.D. Losty).
progress of all children treated at our center focusing on (I) length of hospital stay, (II) readmission rate(s), and (III) need for second intervention(s). Factors potentially related to outcome such as (I) age, (II) comorbidity, and (III) infecting organism(s) were examined in detail. Any feature(s) relating to failure of chemical fibrinolysis was also scrutinized with the aim of identifying patient subgroups unsuitable for fibrinolysis as primary therapy. In September 2006, the seven valent pneumococcal conjugate vaccine (PCV7) was added to routine childhood immunization protocols in the UK [6]. In March 2010 the vaccine was replaced by a thirteen valent pneumococcal conjugate vaccine after public health concerns were raised regarding a growing predominance of invasive serotypes not covered by PCV7 [7]. This study therefore also reports microbiology findings and outcome metrics in the postvaccination era. At our center after a diagnosis of empyema is confirmed by ultrasound all children are then managed by a multidisciplinary (MDT) team led by pediatric respiratory physicians and surgeons. CT chest scans are not a routinely deployed investigation in ‘work up’. Here CT is reserved only for patients where a bronchopleural fistula is suspected or an underlying congenital lung lesion, eg, congenital cystic adenomatoid malformation (CCAM). CT is further used in patient care
http://dx.doi.org/10.1016/j.jpedsurg.2015.07.022 0022-3468/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Long A-M, et al, ‘Less may be best’—Pediatric parapneumonic effusion and empyema management: Lessons from a UK center, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.07.022
2
A-M. Long et al. / Journal of Pediatric Surgery xxx (2015) xxx–xxx
children b10 kg and 40,000 IU diluted in 40 ml of 09.9% saline for children N 10 kg). Primary thoracotomy is reserved for children with a suspicion of bronchopleural fistula(e) and severe necrotizing lung disease with severe systemic illness or in cases with a suspected lung lesion.
Table 1 Population characteristics. Characteristic Sex: Male Female Median age (IQR) Age range Severe comorbidity No Yes ICU/HDU Admission No Yes Causative organism S. pneumoniae Not identified Group A Strep. S. aureus Othersa
n (%) 122 (51) 117 (49) 4 years, (2–8 years) 4 months–19 years 209 (87) 30 (13) 187 (78) 52 (21) 112 (56) 61 (26) 19 (8) 4 (2) 5 (2)
Primary management strategy IV antibiotics alone Chest drain and urokinase Primary thoracotomy Total a b
Failed primary treatment (required second procedure) 66 (27) 169 (71) 4 (2) 239
2b (3) 25 (14.7) 0 27 (11.3)
Mycobacterium tuberculosis 1, Enterococcus faecium 1, Streptococcus intermedius 3. Readmitted and required thoracotomy.
plans when a primary or rescue thoracotomy is considered. Clinically stable patients with small pleural collections seen on ultrasound imaging are therefore managed with antibiotics and active observation. Cefuroxime is employed as a first-line antibiotic and oral clarithromycin added in patients N5 years. Blood cultures, serology, Anti-streptolysin O titre (ASOT), and blood pneumococcal PCR are obtained on all patients at first presentation with pleural fluid routinely sent to the laboratory(s) for microscopy, culture and sensitivity including pneumococcal PCR. Patients with persistently high fever and significant pleural collections are then managed with an 8 Fr chest drain inserted under general anesthesia using the Seldinger technique with instillation of intrapleural urokinase. Six doses of urokinase are administered at 12 hourly intervals (10,000 IU diluted in 10 ml of 0.9% saline was used for
Methods This clinical study was Institutional Review Board (IRB) approved by Alder Hey Childrens Hospital (in accordance with the Declaration of Helsinki 1964). Medical case records were analyzed on patients admitted to our center with a diagnosis of empyema during the era(s) January 2006–December 2012. All patients were managed using the protocol driven MDT care pathway. Data examined were (I) demographics (II), patient comorbidity (III) infecting organism[s], (IV) radiology, and (V) antibiotic therapy. Length of hospital stay, admission to the intensive care unit (ICU) or high dependency (HDU), hospital readmission and need for second intervention(s) were examined. Data on readmission and postprocedural length of stay were fully available from 2007 onward. The influence of severe patient comorbidity(s) and microorganism(s) on hospital stay and admission to ICU or HDU was analyzed with Mann– Whitney U, Chi χ2 and Fishers Exact tests. Factors linked to fibrinolytic failure were identified prospectively: (I) patient age, (II) significant comorbidity, (III) duration of intravenous antibiotic[s] prior to drainage, and (IV) suspicion of necrotizing disease on initial chest imaging. Covariates were selected a priori based on clinical hypotheses. Significant comorbidity included (a) chromosomal anomaly, (b) neurological disability, (c) congenital heart disease, (d) connective tissue disorders, and (e) immune deficiency states. Necrotizing disease on initial imaging was defined as evidence of parenchymal destruction, suppuration or air in the pleural space with chest x-ray, ultrasound or CT. The influence of these factors was further assessed with binary logistic regression to undertake univariate analysis. All statistical analyses were performed using SPSS 20 (with P b 0.05 equals significant). Results A total of 239 patients were admitted to our hospital with empyema in the era(s) 2006–2012. Table 1 shows the demographics with primary management strategies for the cohort (age range 4 months–19 years: median 4 years). The proportion of those who had significant
Fig. 1. Bar chart illustrating empyema hospital admissions per year and management.
Please cite this article as: Long A-M, et al, ‘Less may be best’—Pediatric parapneumonic effusion and empyema management: Lessons from a UK center, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.07.022
A-M. Long et al. / Journal of Pediatric Surgery xxx (2015) xxx–xxx Table 2 Univariate analysis of factors related to failure of fibrinolytic therapy. Factor
Odds ratio
(95% CI)
P value
Age Presence of comorbidity Prior duration of IV antibiotics Suspicion of necrotizing disease on initial imaging
0.96 1.7 0.98 8.69
0.87–1.1 0.54–5.2 0.85–1.12 2.15–35.0
0.51 0.31 0.77 0.002*
comorbidity(s) was 12.6%. First-line therapy involved antibiotics alone in 27% cases; pigtail chest drain and urokinase in 71% with only 2% requiring surgery. Here a primary thoracotomy was undertaken in 4 cases. One patient had an underlying congenital lung lesion that became infected and 3 operations were required emergently for lung abscess/ necrotizing disease with bronchopleural fistulae. The causative microorganisms were identified in 140 patients (59%; Table 1). These included Streptococcus pneumoniae (56%), Group A streptococcus (8%), and Staphylococcus aureus (2%) with a single patient here having MRSA. A declining number of empyema admissions year-on-year was observed with no obvious change in disease severity at first presentation. We did not encounter an increase in the number of cases requiring surgery, ie, early thoracotomy (Fig. 1). Median hospital stay was 10 days [IQR 8–13 days]—with a postprocedural length of stay in those cases requiring intervention of 9 days [IQR 7–12 days]. Patients with significant comorbidity(s) had a significantly longer hospital stay [median 14 days vs median 10 days in otherwise healthy children (P = 0.001)]. Children with comorbidity(s) were significantly more likely also to require ICU or HDU admission—47% vs 18% of otherwise healthy children (P = 0.001). It was also noted that those with pleural disease linked with Group A streptococcus stayed significantly longer in hospital vs cases attributed to S. pneumoniae infection [median 15 days vs 10 days—P b 0.001]. Patients with Group A streptococcal disease also had more severe illness with 58% requiring HDU or ICU admissions compared with 19% cases infected by S. pneumoniae (P = 0.001). Six percent of children were readmitted to hospital (data based on complete analyses from 178 cases). Median time to readmission was 6 days. Of the 11 cases requiring hospital readmission, 5 required thoracotomy—2 of these children had previously been managed with antibiotics alone and 3 with a chest drain and fibrinolysis only. The overall failure rate of the primary treatment modality was 11.3%. Of those who were treated with fibrinolytic therapy as first-line therapy, 25 cases (14.7%) required a second procedure. Twenty patients had a thoracotomy, 2 had thoracoscopic decortication (VATS) and 3 required insertion of a second chest drain We further analyzed the cohort that were treated initially with a chest drain and urokinase and required a second
3
intervention. Factors influencing failure of primary management were assessed with univariate analysis using binary logistic regression (Table 2). The only factor found to be statistically significant and associated with the need for second intervention after fibrinolysis failure was the existence of necrotizing disease suspected from primary radiology imaging (P = 0.002, OR 8.69). There were 14 patients with such findings in the entire study—2 cases were treated initially with primary thoracotomy, 3 with antibiotics alone and 9 patients with a chest drain and urokinase. Of those treated with a chest drain and urokinase, 5 required a second procedure. Three underwent thoracotomy and decortication, a single patient had thoracoscopic debridement (VATS) and insertion of a larger drain and 1 case had a second chest drain inserted. The 5 cases who required a second intervention had a median length of stay of 15 days. In the study one patient died giving an overall mortality rate from thoracic empyema of 0.4% at this center. The single fatality had significant comorbidity—severe neurological disorder with chronic existing lung disease and developed progressive respiratory failure 2 months after treatment of empyema. Discussion This study has highlighted the contemporary outcome(s) of a large cohort of 239 children with pleural empyema managed at a UK pediatric center. All patients were triaged in a consistent way using a MDT protocol-driven care plan. The care pathway illustrates that a ‘less may be best’ policy works for the vast majority of children with early deployment of a fine bore chest drain and fibrinolysis as first line treatment [8]. Outcomes herein reported included factors influencing hospital stay and need for readmission and indication(s) for secondary intervention. The majority of cases were all successfully managed with a chest drain and fibrinolysis with only 14.7% of the cohort requiring a second procedure. These findings compare favorably to several published studies [4,5]. Median postprocedural length of stay was 9 days with a hospital readmission rate of 6.2%. Table 3 provides a summary of outcomes from several published studies including the present Liverpool series. Children with significant comorbidity(s) required an additional median hospital stay of 4 days longer than otherwise previously healthy patients. These cases were also significantly more likely to require ICU or HDU admission. We also found that the infecting microrganism had a significant influence on the likelihood of being admitted to ICU or HDU. Herein the majority of patients presenting with Group A streptococcal infection required ICU admission reflecting a causal association of invasive streptococcal disease with severe systemic toxin mediated illness—‘toxic shock syndrome’ [9]. We observed a falling number of patients admitted to our hospital year-on-year after introduction of the pneumococcal conjugate vaccine with no apparent increase in complexity of disease. This phenomenon
Table 3 Outcome measures from other published studies utilising chemical fibrinolysis. Author (year)
Study type
Protocol
LOS (d)
T/PP
Failure rate (%)
Thomson et al. (2002) [14] Gates et al. (2004) [15] Gates et al. (2004) [16]
RCT SR RR
Drain (both fine-bore and large) + UK Chest Drain + UK or SK Chest Drain and TpA
RCT RCT Database analysis RR RR RR
Drain + UK Drain + TpA Chest drain + fibrinolysis Chest drain + TpA 8 Fr Pigtail + UK 8 Fr Pigtail + UK
PP PP T PP PP PP – PP PP T PP
6.66 N/A N/A
Sonnappa et al. (2006) [4] St Peter et al. (2009) [5] Goldin et al. (2012) [17] Gasior et al. (2013) [18] Petel et al. (2013) [19] Long et al. (2015) (this study)
7.4a 18.9b 8b 6.2b 6b 6.8b N/A 7.2a 11.1b 10b 9b
16.6 16.6 21 15.7 40 14.7
SR, systematic review; RR, retrospective review; RCT, randomized controlled trial; LOS, length of stay; T, total length of stay; PP, postprocedural length of stay; UK, urokinase; SK, streptokinase; TpA, tissue plasminogen activating factor; N/A, not assessed. a Mean. b Median.
Please cite this article as: Long A-M, et al, ‘Less may be best’—Pediatric parapneumonic effusion and empyema management: Lessons from a UK center, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.07.022
4
A-M. Long et al. / Journal of Pediatric Surgery xxx (2015) xxx–xxx
has been noted in other European countries perhaps suggesting some herd protection in those N2 years old who are not routinely offered vaccination [10,11]. S. pneumoniae however remained the most common infecting organism accounting for 56% of the index cases reported in this study. When we analyzed and sought to identify which children were not suitable for chest drainage and urokinase as a first-line treatment for empyema it was only the suspicion of necrotizing disease on initial chest imaging that was associated with an increased treatment failure rate (55% vs 12.4% without). Even so, 45% of this subgroup with the suspicion of necrotizing disease were mostly managed with fibrinolytics and did not require further interventions. To this end, we have not changed our practice to offer such patients who are more severely ill ‘up front’ primary thoracotomy. The information derived from this subgroup analysis is however an important factor to consider particularly when counselling families as the chance of therapy success (%) with chemical fibrinolysis alone is considerably lower versus those children without suspicion of necrotizing disease. We have previously published several report(s) highlighting experience with pediatric empyema some years ago. Evolving practice over twenty years with an MDT team (physicians and surgeons) and the ready availability of a protocol driven care pathway strongly convinces us that ‘less may be best’ for pleural empyema. It is therefore now possible for many children to avoid major surgical operation(s) notably VATS or thoracotomy for empyema [4,5,8,12,13].
References [1] Hamm H, Light RW. Parapneumonic effusion and empyema. Eur Respir J 1997;10: 1150–6. [2] Balfour-Lynn IM, Abrahamson E, Cohen G, et al. BTS guidelines for the management of pleural infection in children. Thorax 2005;60(Suppl. 1):i1–i21.
[3] Islam S, Calkins CM, Goldin AB, et al. APSA outcomes and clinical trials committee. The diagnosis and management of empyema in children: a comprehensive review from the APSA Outcomes and Clinical Trials Committee. J Pediatr Surg 2012;47: 2101–10. [4] Sonnappa S, Cohen G, Owens CM, et al. Comparison of urokinase and video-assisted thoracoscopic surgery for treatment of childhood empyema. Am J Respir Crit Care Med 2006;174:221–7. [5] St Peter SD, Tsao K, Spilde TL, et al. Thoracoscopic decortication versus tube thoracostomy with fibrinolysis for empyema in children: a prospective, randomized trial. J Pediatr Surg 2009;44:106–11. [6] Public Health England. Immunsation against infectious diseases. https://www.gov. uk/government/collections/immunisation-against-infectious-diseasethe-greenbook. [Accessed May 2nd 2014]. [7] Hanquet G, Kissling E, Fenoll A, et al. Pneumococcal serotypes in children in 4 European countries. Emerg Infect Dis 2010;16:1428–39. [8] Khalil BA, Corbett PA, Jones MO, et al. Less is best? The impact of urokinase as the first line management of empyema thoracis. Pediatr Surg Int 2007;23:129–33. [9] Lappin E, Ferguson AJ. Gram-positive toxic shock syndromes. Lancet Infect Dis 2009; 9:281–90. [10] Angoulvant F, Pereira M, Perreaux F, et al. Early impact of 13-valent pneumococcal conjugate vaccine on community-acquired pneumonia in children. Clin Infect Dis 2014;58:918–24. [11] Klugman KP. Editorial commentary: a tale of 2 pneumococcal vaccines. Clin Infect Dis 2014;58(7):925–7. [12] Shankar KR, Kenny SE, Okoye BO, et al. Evolving experience in the management of empyema thoracis. Acta Paediatr 2000;89:417–20. [13] Proesmans M, Gijens B, Van de Wijdeven P, et al. Clinical outcome of parapneumonic empyema in children treated according to a standardized medical treatment. Eur J Pediatr 2014;173:1339–45. [14] Thomson AH, Hull J, Kumar MR, et al. Randomised trial of intrapleural urokinase in the treatment of childhood empyema. Thorax 2002;57:343–7. [15] Gates RL, Caniano DA, Hayes JR, et al. Does VATS provide optimal treatment of empyema in children? A systematic review. J Pediatr Surg 2004;39:381–6. [16] Gates RL, Hogan M, Weinstein S, et al. Drainage, fibrinolytics, or surgery: a comparison of treatment options in pediatric empyema. J Pediatr Surg 2004;39:1638–42. [17] Goldin AB, Parimi C, LaRiviere C, et al. Outcomes associated with type of intervention and timing in complex pediatric empyema. Am J Surg 2012;203:665–7. [18] Gasior AC, Knott EM, Sharp SW, et al. Experience with an evidence-based protocol using fibrinolysis as first line treatment for empyema in children. J Pediatr Surg 2013;48:1312–5. [19] Petel D, Li P, Emil S. Percutaneous pigtail catheter versus tube thoracostomy for pediatric empyema: a comparison of outcomes. Surgery 2013;154:655–60.
Please cite this article as: Long A-M, et al, ‘Less may be best’—Pediatric parapneumonic effusion and empyema management: Lessons from a UK center, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.07.022