Treatment of Empyema in Children: From Hippocrates' Time to the Present, and Back Again

The importance of draining a pleural space infection was known to Hippocrates, and the importance of avoiding complications by definitive treatment at an early stage may have been inferred. As he said with characteristic brevity, "When the empyema is treated either by cautery or incision, if pure and white pus flow from the wound, the patients recover, but if mixed with blood, slimy and fetid, they die" (1).

The treatment of empyema remains controversial, particularly in children. Therapeutic options include systemic antibiotics alone: thoracentesis; tube thoracostomy, with or without instillation of fibrinolytic agents; and more invasive techniques such as thoracoscopic surgery, mini-thoracotomy, and standard thoracotomy with decortication (removal of the fibrinous "peel" from the lungs). How is one to choose among these options and why is this area so controversial? First, experience in adults cannot be extrapolated to children. In contrast with adults, most children who develop empyema are previously healthy. Second, prognostic factors that can help guide invasiveness of therapy in adults, such as pleural fluid lactate dehydrogenase (LDH), glucose levels, and pleural fluid pH, are not as useful in children (2, 3). In addition, outcomes criteria vary widely between studies.

The American College of Chest Physicians published evidence-based guidelines for the management of parapneumonic effusions in adults in 2000 (4). Patients were grouped into four risk categories on the basis of pleural space anatomy, pleural fluid bacteriology, and chemistry, and drainage was recommended for the two highest risk groups. Furthermore, patients in these higher categories had better outcomes (less mortality and failure of primary interventions) when treated by fibrinolytics or surgical intervention than patients treated with drainage alone. These findings are of questionable relevance in pediatrics because risk categories are not so easily defined in children and mortality is fortunately extremely rare. In addition, a subsequent meta-analysis (5) and Cochrane review (6) both tailed to support the routine use of fibrinolytic therapy.

Conflicting data also complicate decision making in children. One randomized controlled trial found a modest decrease in length of stay (7.4 vs. 9.5 d) in children treated with intrapleural urokinase compared with those treated with saline (7) . In contrast, another prospective trial did not find a substantial benefit of fibrinolysis (8). Several studies in children favor a primary operative approach over chest tube alone or chest tube plus fibrinolytics. Gates and colleagues (9) showed shorter length of stay (LOS) and duration of fever in children treated with open thoracotomy or video-assisted thoracoscopic surgery (VATS) compared with chest tube alone or chest tube plus fibrinolytics. Avansino and colleagues (10) showed a 10-fold reduction of failure rate in children treated with operative versus nonoperative therapy. Cohen and colleagues reported a 39% failure rate in children treated with chest tube alone, and a 100% success rate in those treated with VATS (11). Enthusiasm for VATS is tempered, however, by studies showing that even children with severe degrees of pleural thickening usually have normal lung function on long term follow-up after drainage by chest tube alone (12). The recently published British Thoracic Society guidelines for the treatment of pleural space infection (2) in children recommend using fibrinolytic agents in any complicated parapneumonic effusion (thick fluid, fibrous strands on ultrasound) or empyema (overt pus) and holding surgery in reserve for those patients not responding to fibrinolytics.

In this issue of the AJRCCM (pp. 221-227), Sonnappa and colleagues compare two treatments of empyema in children: chest tube plus instilled urokinase versus primary VATS, in a randomized prospective clinical trial, the first of its kind in children (13). The authors chose these two therapies because of the assumed superiority of either therapy over chest tube drainage and antibiotics; however, one should note that neither may be superior for children with early, uncomplicated effusions. No differences were found between the two groups in LOS postintervention, total hospital stay, failure rate, and 6-mo radiologic outcome. Cost was significantly higher in the VATS group, and the authors concluded that chest tube plus urokinase is the preferred treatment. The authors are to he congratulated in designing and carrying out this careful study. One would certainly say that, on a cost basis, urokinase is the better choice if the therapies are equivalent in every other way. However, their recommendation that urokinase is the treatment of choice depends on the clinical equipoise that their study demonstrated.

One concern about this study is that while the groups treated with VATS and fibrinolysis are closely matched, there may have been some small differences. Although the differences in LDH (10,000 in the VATS group and 6,953 in the fibrinolysis group) and prevalence of Streptococcus pneumoniae in the blood or pleural fluid (60% in the VATS group and 43% in the fibrinolysis group) did not reach statistical significance, these findings do point to slightly worse disease in the VATS group, and the study may not have had the statistical power to show if the groups were indeed different in these regards. Pleural fluid white blood cell count, glucose, pH, and bacteriology of the pleural fluid alone were not reported, and may have helped ensure the similarity of the two groups.

A second concern is that LOS after intervention, total LOS, and failure rates are surrogate outcomes, subjectively determined by the attending physician, who is not blinded to the treatment arm. Other important outcomes include the following: the amount of pain associated with the intervention, and long-term pulmonary physiologic outcomes such as pulmonary function and exercise tolerance. These variables are certainly more important than a $2,000 difference in cost.

The above notwithstanding, the study by Sonnappa and colleagues is clearly a helpful step in the direction of providing needed insight into the management of pleural infection in children.

Recently, clinical pathways have been suggested that emphasize individual patient assessment and staging, with prompt treatment and escalation of therapy in nonresponders (2, 9, 14, 15). We favor this approach. Local expertise is an important consideration, and studies that will identify those patients who would benefit from initial surgical therapy are still needed. As Hippocrates said. "He will manage the cure best who has foreseen what is to happen from the present state of matters" (1).

Conflict of Interest Statement: Neither author has a financial relationship with a commercial entity that has an interest in the subject of this manuscript.

© 2006 American Thoracic Society Provided by ProQuest LLC. All Rights Reserved.