Competing risk analysis, as used in AMPLE-2, is a superior approach in this setting, according to Dr Ost. “A plot of the cumulative incidence function would show that initially the groups split apart but the patients with a trapped lung would eventually catch up so the curves would merge,” Dr Ost wrote.3 “However, the curves are not the same so concluding there is no difference would be incorrect.” The competing risk model is more complete because it considers all available information.
In addition, survival analysis techniques more accurately reflect the timing of pleurodesis. Because of differences in the length of time between drainage in the 2 groups in AMPLE-2, pleurodesis would be detected at a maximum of 3 days in the daily drainage group compared with 2 to 4 weeks in the symptom-guided group. With the use of single time point incidence proportions, it would be difficult to determine whether 1 drainage strategy leads to earlier pleurodesis.3
However, with competing risk analysis, one can “quickly inspect the cumulative incidence function curves and see that, even after accounting for up to a 4 week delay in detection, pleurodesis rates favor aggressive drainage,” explained Dr Ost. “The use of survival analysis techniques for studies of malignant pleural effusion should become the standard.”
Dr Ost also noted that patient concerns pertain to freedom from breathlessness and quality of life rather than pleurodesis, and it is important to evaluate the balance between dyspnea relief, pleurodesis, pain, and the influence of these various factors on quality of life. Although an intervention may increase pain, for example, it may also improve quality of life as a result of dyspnea relief. At this time, it appears that “[t]his type of study design, measuring multiple relevant patient-centered outcomes along with a measure of utility tracked longitudinally is an effective method for gaining a more nuanced insight into the merits of competing strategies.”
In practice, however, management strategies should be individualized to each patient. When patients are unsure of their preference regarding drainage strategy, Dr Ost suggests starting with daily drainage for 1 week, followed by the symptomatic approach if drainage does not decrease with the daily approach. This allows patients to choose the strategy that works best for them.
“The goal of shared decision making is optimizing the patient’s quality of life, rather than optimizing the quality of life of a population,” Dr Ost concluded. “Well-designed studies such as this one can inform the decision but require insight when applied at the bedside.”
- Aydin Y, Turkyilmaz A, Intepe YS, Eroglu A. Malignant pleural effusions: appropriate treatment approaches. Eurasian J Med. 2009;41(3):186-193.
- Muruganandan S, Azzopardi M, Fitzgerald DB, et al. Aggressive versus symptomguided drainage of malignant pleural effusion via indwelling pleural catheters (AMPLE2): an openlabel randomised trial. Lancet Respir Med. 2018; 6(9):671-680.
- Ost DE. Quantifying outcomes for palliative pleural interventions. Lancet Respir Med. 2018;6(9):648-649.3.