毛孢子菌屬之流行病學、鑑定方式、抗黴菌藥物敏感性與毛孢子菌血症預後因子之研究

The epidemiology, identification, antifungal susceptibility of Trichosporon species, and predictors of outcome of Trichosporon fungemia

郭欣慧(Shin-Huei Kuo)

指導教授 ： 盧柏樑

高雄醫學大學/醫學院/臨床醫學研究所碩士班/碩士(2019年)

摘要

毛孢子菌屬是一群類酵母黴菌，廣泛存在於環境中，除造成表皮性感染或過敏性肺炎，因侵入性醫療行為或免疫低下罹患侵襲性毛孢子蟲病個案逐年增多，即使給予抗黴菌藥物治療，預後仍舊極差。在台灣，毛孢子菌屬感染症的流行病學資料不多，傳統與自動化鑑定方法的限制，以及其抗黴菌藥物感受性目前仍無臨床判定標準，僅能以in vitro藥物敏感測試結果做為藥物使用依據，這些問題都有待進一步研究。
本研究收集2010年1月1日至2018年12月31日間，高雄醫學大學附設中和紀念醫院、國立台灣大學醫學院附設醫院、中國醫藥大學附設醫院、高雄長庚紀念醫院臨床檢體鑑定為Trichosporon spp.之117株菌株，以Trichosporon ashii最多(85/117, 72.6%)，T. dermatis居次(11.1%)，其次為T. fecales和T. montevideense (6%)。相較於ITS/IGS1 sequencing的精確鑑定，MALDI Biotyper 3.0 (Bruker)可提供精確的種名鑑定，僅無法區分T. dermatis與T. mucoides。
此外，共5種基因型之T. asahii被分離，以第一型最多(41.2%)，接續為第三型(29.4%)；北部與中部地區以第一型居多(45.1%)，其次為第四型(p=0.032)，南部地區則以第三型為主(47.1%, p= 0.004)。以YeastONE broth判定最低抑菌濃度(MIC)，第四型具高amphotericin B MIC(≥ 2 μg/ml)及高fluconazole MIC (≥ 8 μg/ml)之菌株，分別達28.6% (p=0.055)與100% (p=0.344)。
分析51位Trichosporon fungemia患者，預後相關負向因子為qSOFA score (p=0.003) 、血小板低下 (p<0.001)、使用呼吸器(p=0.016)，正向相關因子為感染源控制(中心導管移除) (p<0.001)；扣除co-infection患者後(n=39)，有46.2%接受有效治療，對7天死亡率有顯著正面影響(p=0.025)，但對14天或30天死亡率則無。扣除index date至死亡日小於2天、未能接受有效治療者，該族群(n=41) 7天死亡率相關因子包括qSOFA score、血液腫瘤疾病、血小板低下、有效治療和感染源控制；多變項迴歸分析顯示，感染源控制 (OR 0.054 , 95% CI 0.004-0.697, p=0.025) 與有效治療(OR 0.086, 95% CI 0.008-0.942, p=0.045) 為獨立保護因子；30天死亡率相關因子則僅存感染源控制(OR 0.070, 95% CI 0.009-0.512, p=0.009)。
而對各類抗黴菌藥物的敏感性分析，T. asahii之MIC值皆較non-asahii Trichosporon species高，fungemia菌株中，高fluconazole MIC值 (≥ 8 μg/ml) 比例甚至達80.0%，但對其他azole類尚具良好感受性，voriconazole仍為臨床上之首選藥物。

關鍵字

毛孢子菌屬 ； 鑑定 ； 藥物敏感性 ； 菌血症

並列摘要

Trichosporon species are yeast-like fungi distributed widely in the environment. Except for superficial infection or hypersensitivity pneumonitis, the increasing cases of invasive trichosporosis due to invasive medical procedures or immunocompromised hosts were reported, and the prognosis was extremely poor despite the prescription of antifungal agents. In Taiwan, the data of epidemiology for Trichosporon species was limited. Not only the traditional or automatic methods for its identification showed limitation, there was still no standard criteria for antifungal susceptibility interpretation but based on the results of in vitro susceptibility test for clinical use.
In this study, one hundred and seventeen isolates of Trichosporon species were collected since January 1, 2010, to December 31, 2018, from Kaohsiung Medical University Hospital, National Taiwan University Hospital, China Medical University Hospital, and Kaohsiung Chang Gung Memorial Hospital. T. asahii was the most frequently isolated species (85/117, 72.6%), followed by T. dermatis (11.1%), T. fecales and T. montevideense (each was 6%). Compared to the specific identification of ITS/IGS1 sequencing, only T. dermatis were misidentified as T. mucoides by MALDI Biotyper 3.0 (Bruker).
In addition, five genotypes of T. asahii clinical isolates were identified with a predominance of genotype 1 (41.2%), and 3 (29.4%). Genotype 1 (45.1%) was more prevalent in the northern and middle region of Taiwan, followed by genotype 4 (p =0.032), whereas genotype 3 isolates were predominant in southern Taiwan (47.1%, p=0.004). Depended on the minimal inhibitory concentration (MIC) demonstrated by YeastONE broth, a higher proportion of genotype 4 strains exhibited high MIC to amphotericin B (28.6%, p= 0.055) and fluconazole (100%, p=0.344).
Among the 51 Trichosporon fungemia patients, the analysis revealed that negative outcome is associated with qSOFA score (p=0.003), thrombocytopenia (p<0.001), and mechanical ventilator use (p=0.016), and the only protecting factor is source control (central venous catheter removal) (p<0.001). After excluding the co-infection patients, 42% of the remaining population (n=39) received effective therapy, and it had a positive impact in 7-day mortality (p=0.025) but not in 14-day or 30-day mortality. If excluding those who died less than 48 hours after the index date thus unable to receive the effect therapy, the 7-day mortality of group (n=41) is associated with several prognostic factors, including qSOFA score, hematological disease, thrombocytopenia, effective therapy, and source control. After entering the multivariable logistic regression model, source control (OR 0.054 , 95% CI 0.004-0.697, p=0.025) , and effective therapy (OR 0.086, 95% CI 0.008-0.942, p=0.045) were main protecting factors. Moreover, the only predictor of 30-day mortality was source control (OR 0.070, 95% CI 0.009-0.512, p=0.009).
In general, T. asahii isolates had higher MICs in antifungal susceptibility test than non-asahii Trichosporon isolates among the antifungal agents tested. Within fungemia isolates, even up to 80% were high fluconazole MIC (≥ 8 μg/ml) strains. However, the susceptible rates to other azoles were still fair, and voriconazole remained the most potent antifungal agent in vitro for clinical use.

並列關鍵字

Trichosporon species ； identification ； drug susceptibility ； fungemia

參考文獻

1. Haupt HM, Merz WG, Beschoner WE, Vaughan WP, Seral R. 1983. Colonization and infection with Trichosporon species in the immunocompromised host. J Infect Dis 147:199–203.

2. Walsh TJ. Trichosporonosis. 1989. Infect Dis Clin North Am 3:43–52.

3. Hoy J, Hsu K-C, Rolston K, Hopfer RL, Luna M, Bodey GP. 1986. Trichosporon beigelii infection: a review. Rev Infect Dis 8:959–67.

4. Shimazu K, Ando M, Sakata T, Yoshida K, Araki S. 1984. Hypersensitivity pneumonitis induced by Trichosporon cutaneum. Am Rev Respir Dis 130:407–11.

5. Colombo AL, Padovan ACB, Chaves GM. 2011. Current knowledge of Trichosporon spp. and trichosporonosis. Clin Microbiol Rev 24: 682–700.