Atasever Furkan, Satici Celal, Akyel Resit, Ceran İbrahim Fatih, Azakli Damla, Cayir Kocal Fatma Elif, Sokucu Sinem Nedime
Department of Pulmonology, Yedikule Chest Disease and Thoracic Surgery Training and Research Hospital, Istanbul, Turkey.
Department of Nuclear Medicine, Yedikule Chest Disease and Thoracic Surgery Training and Research Hospital, Istanbul, Turkey.
Respir Med. 2025 Jan;236:107854. doi: 10.1016/j.rmed.2024.107854. Epub 2024 Nov 1.
Assessing the prognosis of patients with malignant pleural effusion (MPE) is cruical to manage the treatment. We aimed to investigate the role of pleural fluid SUVpeak value in predicting mortality.
In our single center, prospective cohort study, we enrolled 158 patients with MPE. PET/CT was independently evaluated by two nuclear medicine physicians. To identify independent predictors of 90-day mortality, we conducted both univariate and multivariate Cox regression analyses. Furthermore, we assessed the discriminatory ability of these predictors compared to the LENT score using the DeLong statistic.
In our study, we enrolled 158 patients with a mean age of 66.29 ± 12.6 years, of whom 100 (63.3 %) were male. The majority of patients (84.9 %) had primary lung cancer, while 24 (15.1 %) had metastases from extrapulmonary malignancies. Among the followed-up patients, 51 (32 %) died within 90 days, with a median survival of 30 (17-61) days. LENT score (HR: 1.52, 95 % CI: 1.14-2.03), pleural fluid albumin value (HR: 0.92, 95 % CI: 0.87-0.97), pleural fluid SUVpeak value (HR: 1.39, 95 % CI: 1.15-1.69), and pleural effusion size (HR: 1.01, 95 % CI: 1.001-1.02) were found to be significant predictors. Furthermore, in assessing the predictive performance for 90-day mortality, ROC curve of a combination of the independent predictors was significantly higher than that of LENT score alone.
Our study revealed that the metabolic characteristics of pleural fluid might predict 90-day mortality in patients with MPE. We established that the incorporation of pleural fluid SUVpeak value into the LENT score has the potential to enhance its discriminative ability.
评估恶性胸腔积液（MPE）患者的预后对于治疗管理至关重要。我们旨在研究胸腔积液SUV峰值在预测死亡率中的作用。
在我们的单中心前瞻性队列研究中，我们纳入了158例MPE患者。PET/CT由两名核医学医师独立评估。为了确定90天死亡率的独立预测因素，我们进行了单变量和多变量Cox回归分析。此外，我们使用DeLong统计量评估了这些预测因素与LENT评分相比的鉴别能力。
在我们的研究中，我们纳入了158例患者，平均年龄为66.29±12.6岁，其中100例（63.3%）为男性。大多数患者（84.9%）患有原发性肺癌，而24例（15.1%）有肺外恶性肿瘤转移。在随访患者中，51例（32%）在90天内死亡，中位生存期为30（17 - 61）天。发现LENT评分（HR：1.52，95%CI：1.14 - 2.03）、胸腔积液白蛋白值（HR：0.92，95%CI：0.87 - 0.97）、胸腔积液SUV峰值（HR：1.39，95%CI：1.15 - 1.69）和胸腔积液大小（HR：1.01，95%CI：1.001 - 1.02）是显著的预测因素。此外，在评估90天死亡率的预测性能时，独立预测因素组合的ROC曲线显著高于单独的LENT评分。
我们的研究表明，胸腔积液的代谢特征可能预测MPE患者的90天死亡率。我们确定将胸腔积液SUV峰值纳入LENT评分有可能提高其鉴别能力。
