胸腔及重症醫學的發炎相關事件

Abstract

背景與目標：敗血症為一因感染導致而危及生命的器官功能障礙，影響數百萬人並有極高的死亡率和合併症，而及早識別感染原因給予正確處置則可以改善治療結果。核子醫學造影，包括鎵-67(Gallium-67)發炎掃描和氟-18去氧葡萄糖 (18F-fluorodeoxyglucose, FDG) 正子掃描 (positron emission tomography, PET)，已廣泛用於識別不明原因發燒的疑似感染疾病。然而，人們對其在重症患者中的作用知之甚少，這些患者通常具有多個發炎病灶並且無法忍受長時間的造影過程。在本研究中，我們旨在評估 FDG PET 對重症患者合併疑似敗血症的診斷效能。 肺部另一個常見的發炎事件是藥物造成的相關發炎。藥物性肝損傷（drug-induced liver injury, DILI）是一個日益重要的問題，特別是在結核（tuberculosis, TB）感染患者的治療中。Statin類藥物可能有肝毒性的危險和statin類藥物在改善某些肝臟疾病預後方面的有益作用是一個相互矛盾的問題。在這項研究中，我們想要評估statin類藥物在抗結核藥物引起的藥物性肝損傷中可能的保護作用。 方法：（1）我們採用系統性回顧和統合分析策略來評估FDG PET和鎵-67掃描對於不明原因敗血症的重症患者是否是良好且安全的診斷影像方式。檢索了截至 2019 年 7 月 24 日的 PubMed 和 Embase，以確定評估 FDG PET 診斷性能的研究，以按照 PRISMA 指南尋找不明原因敗血症之重症患者的感染病灶。以雙變數混合效應模型用於匯集診斷表現的測量值。出版性偏差則是透過 Deeks 方法進行評估。 (2) 我們利用國立台灣大學醫學院附設醫院整合醫療資料庫 (NTUH-iMD) 的整合資料進行了一項以醫院為基礎的回溯式世代研究。結核病培養呈陽性的患者也被納入其中。Statin類藥物的使用定義為每日當量劑量≥0.5毫克pitavastatin。根據是否有肝指數或膽色素升高評估肝功能惡化。主要和次要終點是 DILI 和嚴重 DILI。Statin類藥物的預後價值透過 Kaplan-Meier 分析和 Cox 比例風險模型進行評估。 結果：（1）共納入4篇研究，共87位患者。所有四項研究均評估了 FDG PET。大多數患者具有呼吸衰竭需要機械通氣（76%），或者是休克需要血管升壓藥物（61%）。與檢查和運送過程相關的不良事件很少見（2%）。統合分析後的敏感度和特異度分別為 0.94 (百分之九十五信賴區間, 0.79–0.99) 和 0.66 (百分之九十五信賴區間, 0.45–0.83)。Summary ROC 曲線的 AUC 為 0.83。 (2)共納入1312例確診結核病並接受抗結核治療的患者。在結核治療期間，193 名患者發生 DILI，140 名患者發生嚴重 DILI。 Kaplan-Meier 分析顯示，DILI 中規則statin類藥物使用者和對照組之間有顯著差異。在多變量 Cox 比例風險分析中，statin類藥物顯示出對主要和次要終點的保護作用。此外，statin類藥物的保護作用對 DILI 表現出劑量反應關係。 結論：（1）FDG PET 是一種非常敏感的工具，具有可接受的特異性，可用於檢測重症患者的感染源。然而，目前可用的研究在評估安全問題方面有其限制。進一步的研究應該調查對這組弱勢患者進行這項測試的益處和風險。 (2) statin類藥物治療對抗結核藥物性肝損傷風險具有保護作用，且呈現正向劑量反應關係。

Background and Objectives: Sepsis is a life-threatening organ dysfunction caused by infection, affecting millions of people with high mortality and complications. Early identification of the infection source and appropriate management can improve outcomes. Nuclear medicine imaging, including Gallium-67 (Ga-67) inflammation scans and 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET), has been widely used to identify suspected infectious diseases in cases of fever of unknown origin. Yet, the role of FDG PET in critically ill patients, who often have multiple inflammatory foci and cannot tolerate prolonged imaging procedures, is less understood. In this study, we aim to evaluate the diagnostic efficacy of FDG PET in critically ill patients with suspected sepsis. Another common inflammatory event in the lungs is drug-induced inflammation. Drug-induced liver injury (DILI) is an increasingly important issue, particularly in the treatment of patients with tuberculosis (TB). The potential hepatotoxicity of statins and their beneficial effects in improving the prognosis of certain liver diseases is a conflicting issue. In this study, we aim to evaluate the potential protective effect of statins on drug-induced liver injury caused by anti-tuberculosis medications. Methods: (1) We adopted a systematic review and meta-analysis strategy to evaluate whether FDG PET and Ga-67 scans are effective and safe diagnostic imaging methods for critically ill patients with sepsis of unknown origin. We searched PubMed and Embase up to July 24, 2019, to identify studies evaluating the diagnostic performance of FDG PET in detecting infectious foci in critically ill patients with unknown sepsis, following PRISMA guidelines. A bivariate mixed-effects model was used to pool measures of diagnostic performance. Publication bias was assessed using Deeks' method. (2) We conducted a hospital-based retrospective cohort study using integrated data from the National Taiwan University Hospital Integrated Medical Database (NTUH-iMD). Patients with positive TB cultures were included. Statin use was defined as a daily equivalent dose of ≥0.5 mg pitavastatin. Liver function deterioration was assessed based on elevated liver enzymes or bilirubin levels. The primary endpoint was the DILI and the secondary endpoint was the severe DILI. The prognostic value of statin use was evaluated using Kaplan-Meier analysis and Cox proportional hazards models. Results: (1) Four studies with a total of 87 patients were included. All four studies evaluated FDG PET. Most patients had respiratory failure requiring mechanical ventilation (76%) or shock requiring vasopressors (61%). There were minimal adverse events related to the examination and transport process (2%). The pooled sensitivity and specificity were 0.94 (95% confidence interval of 0.79–0.99) and 0.66 (95% confidence interval of 0.45–0.83), respectively. The area under the summary ROC curve was 0.83. (2) A total of 1,312 patients diagnosed with TB and undergoing anti-tuberculosis treatment were included. Among them, 193 patients experienced the DILI during the course of TB treatment, with 140 of them developing the severe DILI. Analysis using the Kaplan-Meier method showed a significant difference in the DILI between regular statin users and the reference group. In multivariate Cox proportional hazards analysis, statin use showed a protective effect on both primary and secondary endpoints. Additionally, the protective effect of statins on DILI exhibited a dose-response relationship. Conclusions: (1) FDG PET is a highly sensitive tool with acceptable specificity for detecting infectious sources in critically ill patients. However, the currently available studies have limitations in evaluating safety issues. Further research should investigate the benefits and risks of this test in this vulnerable patient group. (2) statin therapy has a protective effect against drug-induced liver injury from anti-tuberculosis medications, demonstrating a positive dose-response relationship.