Article Text

Empyema thoracis: a role for open thoracotomy and decortication
  1. JEREMY HULL, Clinical Lecturer in Paediatrics,
  1. Department of Paediatrics, John Radcliffe Hospital
  2. Headington, Oxford OX3 9DU, UK
    1. ANNE THOMSON, Consultant in Respiratory Paediatrics
    1. Department of Paediatrics, John Radcliffe Hospital
    2. Headington, Oxford OX3 9DU, UK

      Statistics from

      Editor,—We were disappointed that Careyet al’s recent article on surgical management of empyema thoracis1 was not balanced by a commentary providing the contrary view that surgery is rarely, if ever, needed in the modern management of empyema. Careyet al described the outcome of 18 children with empyema treated by open thoracotomy and decortication. The outcome in all 18 was good, with a mean length of stay in hospital after surgery of four days. They rightly emphasise that this treatment needed to be carried out in a regional cardiothoracic centre with experienced paediatric anaesthetists and surgeons and a paediatric intensive care unit, where several of the “younger” patients in this study spent the night still intubated. This treatment is then not readily available to most children who develop empyema.

      We agree that, in expert hands, surgery can be an excellent treatment for empyema, however, we do not believe that this report advances the current management of this condition. The authors make no attempt to compare open surgery with other treatments, such as closed drainage and fibrinolysis (with urokinase or streptokinase), and admit that there are no controlled trials of different management approaches for empyema. Therefore there are no data, either in their paper or in the available literature, to support their conclusion that “even in ideal cases, neither fibrinolysis nor thorascopic adhesiolysis can achieve more rapid resolution at lower risk [than open surgery]”.

      Our experience of closed drainage (pigtail catheter) and urokinase treatment given 12 hourly for three days for all children referred to our centre with empyema (n = 27 over two years; 20 with loculated empyema on ultrasound) is that it results in rapid resolution of fever and early discharge, with a mean length of stay of five days. Since we have been using urokinase, none of the children treated has required surgery. The choice therefore would appear to be between a general anaesthetic, an operation, a possible intensive care unit stay, and a scar, or sedation, a soft chest tube, no scar, and an equally good outcome.

      To determine whether fibrinolysis is as effective as it seems, we have embarked on a multicentre, double blind, randomised controlled trial (11 UK centres) of closed drainage with urokinase or saline in the management of empyema. One of the end points in this study is a requirement for surgery due to failure of resolution of symptoms. We encourage paediatricians looking after children with empyema to consider entering the study.


      Dr Carey comments:

      Lack of facilities cannot be used as an argument against surgical treatment. Referral to facilities with the required expertise is possible throughout the United Kingdom. Surgeons and anaesthetists should not undertake occasional paediatric practice.1-1

      We stand by our statement that there is no evidence to show that fibrinolysis or thoracoscopic adhesiolysis can achieve a better or safer outcome than that presented in our surgical series of stage II and stage III empyema. As we have already stated, alternatives to surgery should be considered in selected cases and progression should be arrested by timely intervention.

      It is difficult to interpret Dr Thomson’s unpublished observations. She has not described how she came to the diagnosis of empyema in these children. She must also define loculation on ultrasound. This term might encompass any stage in empyema progression including fine fibrinous strands wafting in a fluid filled pleural space, fibrinopurulent material with multiloculated pus, or densely organised areas interspersed with fibrinopurulent material or fluid pockets. This classification also ignores the area of pleural space involvement, and the extent of rind formation and lung entrapment. We encourage prospective randomised trials but any meaningful study must accurately stage and stratify the extent of disease.

      Ultrasound is a reasonable screening tool to stage the disease at presentation and monitor disease progression. However, it is not accurate enough to stratify disease severity for the purpose of such a trial. The findings at operation correlate poorly with the preoperative ultrasound report, particularly in relation to the presence and extent of rind formation and drainability of “fluid”, which turns out to be thick fibrinopurulent material undergoing organisation.

      The fact that Dr Thomson could even contemplate saline treatment is indicative of the very early stage of empyema that she is fortunate to see in her practice. We have commented on the delays in surgical referral in our series. If more advanced disease is encountered, it would not be appropriate to compare surgery and fibrinolysis until fibrinolysis has been proved to be effective in children.

      Finally, we emphasise that the choice between general anaesthetic, operation, and scar versus sedation and soft chest tube is dictated by disease stage and extent and, hence, time of referral. Chest drain insertion under sedation is traumatic. General anaesthetic is necessary for rigid bronchoscopy and accurate drain placement in advanced disease.


      1. 1-1.

      Request Permissions

      If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.