Description of the condition
Pleural empyema refers to pus in the pleural cavity and is a sequelae of bacterial seeding of the pleural space. Possible causes include complications from pneumonia, penetrating chest wall injury and thoracic procedures (Ferguson 1996). Pleural empyema is most commonly the result of pneumonia, with 20% to 57% of patients with pneumonia invariably developing parapneumonic effusion which can progress to pleural empyema (Sahn 2007). Parapneumonic effusion may resolve without further complications or progress to pleural empyema. The stages of progression are as follows: the initial exudative stage is characterised by a shift of pulmonary interstitial fluid into the pleural space from an increased capillary permeability. The second stage is the fibropurulent stage, where the pleural space becomes infected and loculation occurs. Lastly, organisation of the chronic inflammation results in proliferation of fibroblasts and a thickening of the pleural space, known as a pleural peel, can be seen on imaging (Light 2006). Clinically, the three stages can be referred to as uncomplicated parapneumonic effusion (UPPE), complicated parapneumonic effusion (CPPE) and pleural empyema respectively. UPPE usually resolves with antibiotic therapy alone but CPPE and empyema require additional interventions. The diagnosis of CPPE requires either positive pleural fluid biochemistry (pH < 7.2, glucose < 2.2 mM/L, lactate dehydrogenase (LDH) > 1000 IU/L) (Appendix 1), suggestion of pleural empyema on medical imaging examinations such as ultrasound or computed tomography (CT), or definitive diagnosis of positive culture or gram stain from pleural aspirate (Sahn 2007). This review will examine the available evidence for the optimal management strategy in the treatment of CPPE and pleural empyema. We will compare surgical options and minimally invasive non-surgical procedures and note potential complications from the different procedures.
Description of the intervention
Both surgical and non-surgical options are available for the management of pleural empyema. Surgical interventions include video-assisted thoracoscopic surgery (VATS) with adhesiolysis or open thoracotomy and decortication of the pleural space (Appendix 1). Non-surgical, minimally invasive procedures include thoracentesis and chest tube drainage, with or without the use of intrapleural fibrinolytics (Appendix 1). A description of these interventions is provided below. Both patient groups will be treated equally apart from the intervention, which means that patients will be on targeted antibiotic therapy based on microscopy, culture and sensitivity (Davies 2010).
Currently there is no consensus as to which patient groups will benefit from surgical intervention over management by minimally invasive procedures. VATS (Appendix 1) enables visualisation of the pleural cavity for drainage of pus and disruption of septations. A temporary chest tube is left in place for postoperative drainage of any re-accumulated effusions (Light 2006). Open thoracotomy involves making an incision into the chest wall to gain direct access to the chest cavity. This enables complete drainage of the empyema or CPPE and direct decortication (Light 2006). Complications of both VATS and thoracotomy include postoperative pneumothorax, intercostal neuralgia and anaesthetic risks (Yim 1996). There is potentially a higher risk of postoperative complications associated with open thoracotomy as it involves a much larger incision and longer operation (Jaffé 2003), hence VATS is the more commonly performed procedure in the surgical management of empyema. Nonetheless, conversion to open thoracotomy may be required (Lardinois 2005).
Procedural management of pleural empyema involves thoracentesis and chest tube drainage. Thoracentesis involves aspirating the pleural fluid through a catheter under imaging guidance of either ultrasound or CT. Potential complications from this procedure include haemothorax, pneumothorax and lung puncture (bronchopleural fistula) (Jones 2003). Alternatively, chest tube insertion involves dissection of a small area of the chest wall muscle and the placement of a chest tube (Oddel 1994). The length of treatment is usually no longer than 7 to 10 days or when drainage is minimal, as guided by clinical or radiographic evidence of empyema resolution (or both) (Oddel 1994). In patients not responding to treatment or requiring a prolonged period of chest tube placement, surgical intervention may be considered. Complications associated with tube thoracostomy include chest tube malposition, tissue trauma and re-expansion pulmonary oedema (Miller 1987).
Intrapleural fibrinolysis uses a mixture of streptokinase and streptodornase and is an adjunct to chest tube drainage to facilitate fibrinolysis of loculations (Tillett 1951). Side effects due to impurities led to a decline in its use but the availability of a more purified form and successful trial of urokinase have led to a reappraisal of this modality (Aye 1991; Temes 1996). More recently, a 2008 Cochrane Review by Cameron et al concluded that intrapleural fibrinolytic therapy conferred significant benefit in reducing the requirement for surgical interventions. However, when subgroup analysis was performed on high-quality trials, the benefit was not significant (Cameron 2008).
Why it is important to do this review
There are various surgical and non-surgical options for the treatment of CPPE and pleural empyema currently used in practice, yet there is no clear consensus on optimal methods for intervention. Coote's 2005 review of the topic concluded that there were insufficient large trials to suggest the use of any particular intervention (Coote 2005). Our review aims to reconcile the issue by comparing the results of surgical and non-surgical therapies for the treatment of CPPE and pleural empyema, thereby providing clinicians with evidence-based guidance for its management.
To assess the validity of surgical versus procedural interventions for CPPE or pleural empyema.
Criteria for considering studies for this review
Types of studies
Randomised controlled trials (RCTs) and quasi-RCTs comparing any surgical intervention to any non-surgical intervention for the management of complicated parapneumonic effusion or pleural empyema.
Types of participants
Participants of all ages and either gender with a diagnosis of CPPE or empyema. Analysis of pleural aspirates or medical imaging examinations such as ultrasound or CT can be used to confirm the diagnosis of CPPE or empyema. The diagnosis of CPPE requires either positive pleural fluid biochemistry (pH < 7.2, glucose < 2.2 mM/L, LDH > 1000 IU/L), confirmation of loculation on imaging examination, or positive culture or gram stain from pleural aspirate. Pus aspirated from the pleural space is indicative of pleural empyema.
Exclusion criteria will be any contraindications to either surgery, minimally invasive procedures or to the fibrinolytic agent. Tuberculous empyemas are excluded because of the different clinical presentation, severe associated complications and poorer outcomes (Chapman 2004; Kundu 2010). Patients who are immunocompromised or have underlying malignancy will also be excluded (Kaifi 2012). Patients with comorbid conditions that would require hospitalisation beyond the course of the empyema will be excluded.
Types of interventions
- Video-assisted thoracoscopy with adhesiolysis (VATS)
- Open thoracotomy and decortication
Comparator: procedural management
- Chest tube insertion
Interventions with or without intrapleural fibrinolytics will be included in the review.
Types of outcome measures
- Length of hospital stay (days)
- Procedural complications
Search methods for identification of studies
We will search the Cochrane Central Register of Controlled Trials (CENTRAL) (current issue), which includes the Cochrane Acute Respiratory Infections Group's Specialised Register, MEDLINE (1950 to current date), EMBASE (1974 to current date), CINAHL (1981 to current date) and LILACS (1982 to current date).
See Appendix 2 for details of the MEDLINE search strategy. We will adapt this search strategy for the other electronic databases. There will be no language, publication year or publication status restrictions on searching.
Searching other resources
We will manually search the reference lists of identified publications for additional trials, either published or unpublished, and contact trial authors if necessary. We will search ClinicalTrials.gov and other trials registers for any ongoing trials or trials which may have been published and were missed.
Data collection and analysis
Selection of studies
Two review authors (TC, MR) will independently assess the studies obtained from the searches to determine eligibility. We will assess the full text of a study if abstracts are unavailable. Two review authors (TC, MR) will independently assess the retrieved trials for compliance with the inclusion and exclusion criteria. A third author (MVD) will resolve any disagreements by acting as an arbiter.
Data extraction and management
We will design a standardised data extraction sheet. Two review authors (TC, MR) will independently extract data. A third author (MVD) will act as an arbiter if any disagreements occur. Two review authors (TC, MR) will enter data into RevMan 2012. All review authors will discuss any discrepancies. One review author (MVD) will act as arbiter.
Assessment of risk of bias in included studies
Two authors (TC, MR) will independently assess included studies for risk of bias using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We will assess the following domains for each study and deem them to be low risk, high risk or unclear risk.
1. Random sequence generation
- Low risk: allocation was generated by a computer or random number table algorithm.
- High risk: any non-random sequence generation. For example, dates, names or identification numbers.
- Unclear risk: sequence generation not mentioned.
2. Allocation concealment
- Low risk: the process used a telephone or central allocation system or sealed opaque envelopes to prevent patient recruiters, investigators and participants from knowing the intervention allocation.
- High risk: other methods of allocation concealment. For example, open random allocation, unsealed or non-opaque envelopes, date of birth.
- Unclear risk: method of allocation concealment not mentioned.
3. Blinding of participants and outcome assessors
- Low risk: blinding was performed adequately, or the outcome measurement is not likely to be influenced by lack of blinding.
- High risk: no blinding or incomplete blinding, and the outcome or the outcome measurement is likely to be influenced by lack of blinding.
- Unclear risk: there is insufficient information to assess whether the type of blinding used is likely to induce bias in the estimate of effect.
4. Incomplete outcome data
- Low risk: there are no missing outcome data, the reasons for missing outcome data were unlikely to be related to a true outcome, or the missing outcome data were balanced in numbers or reasons across intervention groups.
- High risk: there are missing outcome data that are likely to be related to the true outcome or there is an imbalance in numbers or reasons for missing data across intervention groups.
- Unclear risk: there are incomplete outcome data that are not addressed.
5. Selective reporting
- Low risk: the trial protocol is available and all of the trial's pre-specified outcomes that are of interest in the review have been reported.
- High risk: not all of the pre-specified primary outcomes of the trial have been reported.
- Unclear risk: there is insufficient information to assess whether the magnitude and direction of the observed effect is related to selective outcome reporting.
6. Other bias
For each study we will describe any concerns we have about other possible sources of bias, for example, financial disclosures, conflicts of interest of authors, availability of ethics approval for the study, or the obtaining of informed consent from participants. We will assess whether each study was free of other problems that could put it at risk of bias as low risk, high risk or unclear risk.
Measures of treatment effect
When dealing with dichotomous outcome measures, we will calculate a pooled estimate of the treatment effect for each outcome across trials using the odds ratio (OR) (the odds of an outcome among treatment-allocated participants compared to the corresponding odds among controls) and the 95% confidence interval (CI). For continuous outcomes, we plan to record either mean change from baseline for each group or mean post-intervention values and standard deviations (SD) for each group. Where appropriate, we then plan to calculate a pooled estimate of treatment effect by calculating the mean difference (MD) and 95% CI.
Unit of analysis issues
The unit of analysis will be each patient recruited into the trials. We will adjust for the cluster effect as described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011) if the level of analysis is not the same as the level of randomisation, such as in cluster-randomised controlled trials.
Dealing with missing data
We will contact the trial authors for missing data when possible. Otherwise we will deal with missing data by performing an intention-to-treat (ITT) analysis that considers all missing data as unsuccessful outcomes.
Assessment of heterogeneity
We will assess heterogeneity based on face value (for example population, setting, risk of complications) and with the I
- 0% to 40% might not be important;
- 30% to 60% may represent moderate heterogeneity;
- 50% to 90% may represent substantial heterogeneity;
- 75% to 100% represents considerable heterogeneity.
The observed importance of the I
Assessment of reporting biases
We will assess reporting bias for each outcome by using funnel plots if there are 10 or more studies. Asymmetrical funnel plots can indicate outcome reporting bias or heterogeneity. If outcome reporting bias is suspected, we will contact study authors. Outcome reporting bias can be assessed by comparing the methods section of a published trial to the results section where the original protocol is not available.
We will perform a meta-analysis if it is appropriate by using RevMan 2012. We will describe studies without pooling them if there is obvious face value heterogeneity. We will use the fixed-effect model if there is no statistical heterogeneity. We will use the random-effects model if there is statistical heterogeneity (Higgins 2011).
Subgroup analysis and investigation of heterogeneity
We will perform subgroup analysis for children and adults, with children defined as those under the age of 18, and adults as those age 18 or over, at the time of diagnosis. This is due to the lower mortality rate of paediatric empyema, compared to adult empyema, which is estimated at 20% (Balfour-Lynn 2005; Ferguson 1996).
We will perform a subgroup analysis for trials involving the use of intrapleural fibrinolytic therapy, as studies have demonstrated an improved outcome when included in empyema therapy.
We will pool studies with a low risk of bias and gradually add studies with a high risk of bias to explore how that changes the overall estimate of effect. We will also explore which studies contribute to heterogeneity by gradually adding studies to the pooled analysis.
We would like to acknowledge the previous authors of this review, Nicky Coote and Elspeth Kay, and thank Sarah Thorning, the Trials Search Co-ordinator for the Cochrane Acute Respiratory Infections Group for designing the search strategy of our review. We also wish to thank the following people for commenting on the draft of this protocol: Amy Zelmer, Mario Kopljar, Craig Mellis, Nelcy Rodrigues and Meenu Singh
Appendix 1. Glossary of terms
Decortication: surgical removal of fibrous tissue in the pleural space.
Thoracentesis: bedside procedure involving a fine needle (usually 12 to 14 G cannula) connected to a syringe that is inserted into the pleural space to drain any fluid for therapeutic or diagnostic purposes.
Intrapleural fibrinolytics: therapy involving administration of fibrinolytics (streptokinase or urokinase) through a chest tube drain to facilitate breakdown of any fibrous loculations and improve the drainage of the pleural collections.
Open thoracotomy: surgical procedure to open up the chest cavity (typically a posterolateral incision in an intercostal space) that is widened by rib spreaders (retractors), to gain direct visualisation and access to the pleural space.
Thoracostomy: bedside procedure involving a small incision between two ribs and the insertion of a flexible plastic tube into the pleural space to drain pleural collections.
Video-assisted thoracoscopic surgery (VATS): surgical procedure involving the insertion of a small video camera along with other surgical instruments through three to four small incision ports into the patient's chest, enabling the chest cavity to be visualised for surgical procedures to be performed.
Mediastinum: a group of structures in the middle of the thorax, including the heart and its surrounding structures.
Dyspnoea: symptom of breathlessness.
Fibropurulent: also known as fibrinopurulent. It pertains to pus or suppurative exudate that contains a relatively large amount of fibrin.
Lactate Dehydrogenase (LDH): An enzyme that functions in anaerobic glucose metabolism and glucose synthesis.
Appendix 2. MEDLINE search strategy
1 exp Empyema, Pleural/
2 exp Pleural Effusion/
3 ((empyema or effusion*) adj3 pleura*).tw.
4 ((purulent or suppurative) adj3 pleurisy).tw.
5 (empyema adj3 thora*).tw.
7 ((parapneumon* or para-pneumon*) adj2 effusion*).tw.
We will combine the MEDLINE search strategy with the Cochrane Highly Sensitive Search Strategy for identifying randomised trials in MEDLINE (Lefebvre 2011).
Contributions of authors
Tze Yang Chin and Mark Redden wrote the protocol under the guidance of Mieke van Driel and Charlie Chia-Tsong Hsu.
Declarations of interest
The authors have no conflict of interest to declare.