2001~2003年台灣南部地區登革熱/登革出血熱的流行病學探討

Epidemiology of Dengue Fever/ Dengue Hemorrhagic Fever in Southern Taiwan, 2001~2003

吳民惠(Min-Huei Wu)

指導教授 ： 金傳春

國立臺灣大學/公共衛生學院/流行病學研究所/碩士(2004年)

https://doi.org/10.6342/NTU.2004.01035

摘要

2002年台灣爆發近六十年最嚴重的一次登革出血熱流行，重症病例數創歷年新高，且登革出血熱的致死率高達8.7%，此波流行病學的重要性值得深入探討。本研究目的是藉由發現2001~2003年台灣高雄縣/市、屏東縣/市之登革出血熱爆發流行時的重要流行病學特徵，進而探討致禍的第二型登革病毒在分子層次上所產生的變異。全文分為三部分：(一)利用疫情調查與學術單位研判的登革熱與登革出血熱病例資料，分析2001~2003年南台灣登革出血熱爆發之流行病學、病毒學及血清學數據；(二)台灣第二型登革病毒全長外套膜(envelope)基因1485核苷酸序列的歷年趨變分析；及(三)比較此病毒分子變異群(quasispecies)在宿主有無潛在疾病或系統性疾病之2001~2002年登革病人血漿/周邊血液單核細胞中的變化差異何在。
2001與2002~2003年兩波流行的登革出血熱病例多出現在流行高峰期(各為2001年10月20日~11月23日的61.5%與2002年9月1日~11月2日的50.8%)。若單比較2002年登革出血熱與登革熱病人兩者之間的危險因子，發現：(1)最開始的流行區為高雄市前鎮區之登革出血熱病例佔全部登革病例的百分比，除樣本數較小的第一期，第二時期起由7.05%(28/397)上升至第三時期的9.7%(14/144)再升至第四期的10.9%(43/394)，且具統計顯著意義(p=0.0003)；(2)登革出血熱病人較常出現在四十至六十歲的中年者[44.79% (118/295)，p=0.008]，具有氣喘[3.04% (7/230) vs. 1.12% (8/715)，p=0.04]、具有腎病[6.52% (15/230) vs. 3.08% (22/715)，p=0.01]、具有高血壓[27.83% (64/230) vs. 18.23% (131/715)，p=0.001]、具有B型肝炎[6.52% (15/230) vs. 3.36% (24/715)，p=0.04]、糖尿病[22.17% (51/230) vs. 14.4% (103/715)，p=0.005]、及胃潰瘍[10.43% (24/230) vs. 4.2% (30/715)，p=0.0004]；(3)登革出血熱的病人主訴與其過去有登革病史具統計顯著相關[18.29% (47/257) vs. 12.16% (300/2467)，p=0.005]，但由血清學診斷卻發現過去有登革病史與是否為登革出血熱尚無統計顯著相關[93.38% (127/136) vs. 88.38% (1148/1299)，p=0.08]。
在分子流行病學方面，2001至2003年登革病人血中分離出的第二型登革病毒30株進行E基因序列全長的比對，發現其有高達99.1%~100%的同質性，仔細分析知2001年的8株、2002年20株與2003年兩株的相似度各為99.6%~100%、99.5%~100%及99.7%。又以1981年的第二型登革病毒株之E基因序列為基準，相較於1987、1997、2001~2003年所得的此型病毒E基因序列之相似度由95.3%已下降至各為94.4%與92.6%。由演化樹比較發現：(1)台灣的第二型登革病毒共有兩群(0%~8.7%變異)，分別為都會型基因型別與亞洲第二基因型別；(2)2001與2002~2003年可以清楚區分兩個明顯的分支，在分子層次上的核苷酸位置137、291、1128上由2001年8株的C、A、T，至2002-2003年的22株已被取代成T(其中一株仍維持C)、T、C，而氨基酸位置46是由2001年的酥胺酸(T)被2002年取代為異白胺酸(I)，即隨著流行時間拉長而基因序列上有一致性變化；(3)有趣的是2001年與2002年均可看出自屏東縣東港鎮的第二型登革病毒株是和其他高雄縣/市、屏東市所分離的第二型登革病毒株有較遠的親源關係，顯示環境的不同與流行的密度會影響病毒演化的速率，或是在不同的選擇壓力下易有不同的演化方向；(4) 在四株境外移入的第二型登革病毒中，2001年自菲律賓感染的第二型登革病毒與2001~2003年台灣本土的第二型登革病毒有最高達98.9%~99%的相似性，經演化樹分析發現此登革病毒也屬於都會型(cosmopolitan)基因型別，而1998年自泰國、2002年自緬甸、2003年自越南境外移入的分離的第二型登革病毒株均屬於亞洲第二基因型別；及(5)疾病嚴重程度(登革出血熱與登革熱)、身體是否有系統性疾病等潛在病因，如高血壓、糖尿病、心臟病，均未在演化樹上呈現獨特而不同的群組。
為了明瞭登革病毒是否在免疫抑制的病人較易產生具有差異的登革病毒變異株，特別選擇了流行病學特徵與病史清楚的4位登革熱與4位登革出血熱病人的血液檢體(七位為血漿，一位為單核球細胞)，進行第二型登革病毒的分子基因群變異研究。利用反轉錄-聚合酶鏈鎖反應及克隆序列(clonal sequencing)，針對外套膜蛋白分別挑選12~20個克隆進行序列分析，結果顯示登革病毒在病人血液檢體中的病毒株群(quasispecies)變異的程度與疾病的嚴重度無關，但是在有潛在病因的登革病人身上有較健康人高的病毒株群變異(p=0.072，

關鍵字

登革熱 ； 登革出血熱 ； 流行病學

並列摘要

The largest epidemic of dengue hemorrhagic fever (DHF) with the highest case numbers and case fatality rate (CFR：8.7%) of DHF caused by dengue virus serotype 2 (DEN-2) exploded in southern Taiwan in 2002, since 1942. We investigated the molecular changes of viral factors plus host variables in this epidemic in KaoHsiung City/County and Pingtung City/County during 2001-2003 by involving three parts: (1) to understand the epidemiological, virological and serological data on DHF versus dengue fever (DF) cases which were defined by dengue team scholars through field investigations, and (2) to elucidate the molecular changes in nucleotide and amino acid sequences of complete envelope (E) gene (1485 nucleotides) of DEN-2 by collecting isolates obtained from 2001-2003 compared with those from past epidemics in Taiwan and other countries from GenBank through phylogenetic and qualitative analyses, and (3) to search for the relationship between the diversity of quasispecies in E gene in DF versus DHF patients with or without different underlying diseases.
The most DHF cases were emerged from the peak of the epidemic curve (35th-44th wk). By univariate analysis of several risk factors for DHF vs. DF cases, our results indicated that : (1)the percentage of DHF cases in ChienJen District of KaoHsiun City (initial epi-center) increased significantly from the second period [7.05%(28/397)] to the third period [9.7%(14/144)] and the fourth period [10.9%(43/394)] (p=0.0003) besides the initial epidemic period; (2) DHF cases occurred more frequently in 40~60 y/o adults [44.79% (118/295)，p=0.008], in patients with chronic diseases such as asthma [3.04% (7/230) vs. 1.12% (8/715)，p=0.04], kidney disease [6.52% (15/230) vs. 3.08% (22/715)，p=0.01], hypertension [27.83% (64/230) vs. 18.23% (131/715)，p=0.001], HBV infection [6.52% (15/230) vs. 3.36% (24/715)，p=0.04], diabetes mellitus [22.17% (51/230) vs. 14.4% (103/715)，p=0.005], gastric ulcer [10.43% (24/230) vs. 4.2% (30/715)，p=0.0004]; and (3) DHF had significant association with the patient’s self-described past dengue history [18.29% (47/257) vs. 12.16% (300/2467)，p=0.005] but smaller-scale serological data did not show such an association with past dengue virus infection [93.38% (127/136) vs. 88.38% (1148/1299)，p=0.08].
Molecular epidemiologic study to determine the role of viral evolution in emerging a large-scale epidemic of DHF during 2001-2003 was to analyze the complete sequences of E gene of 38 DEN-2 isolates obtained from epidemics in 1981, 1987, 1997, and 2001-2003 in Taiwan. The maximum-likelihood phylogenetic tree analysis revealed that Taiwan’s DEN-2 isolates fell into 2 clusters (diversity 0~8.7%). The 2001-2003’s DEN-2 isolates, which belonged to the cosmopolitan genotype, showed 99.1%~100% sequence identity and 8, 20 and 2 DEN-2 isolates of 2001, 2002 and 2003 had 99.6-100%, 99.5-100% and.99.7% homology, respectively. The 1981’s DEN-2 isolate had 95.3% nt homology with 1987’s and such an identity dropped to 94.4% in 1997, and declined to 92.6% in 2001-2003. The nucleotide sequences of E gene at positions of 137, 291, 1128 of 2001’s eight DEN-2 isolates had consistent changes from C, A, T to T (only one remained as C), T, C in 2002-2003’s twenty-two isolates whereas the amino acid at position 46 was consistently changed from Thr in 2001 to Ile in 2002 as the epidemic became longer. Tungkang’s DEN-2 isolates showed geographic differences from Kaoshoung’s DEN-2 isolates in both 2001 and 2002, implying different environmental factors or selective pressures affecting various evolution rate or directions of DEN-2. Among 4 imported DEN-2 isolates, the one from the Philippines (Cosmopolitan genotype) had the highest homology (98.9-99%) with Taiwan’s indigenous DEN-2 isolates in 2001-2003. There were no molecular signatures of distinct lineage for those isolates from DHF vs. DF or dengue cases with or without specific underlying diseases.
To understand whether quasispecies of DEN-2 would have more variation in those dengue patients with underlying diseases (diabetes, hypertension etc.), 7 plasma and 1 peripheral blood mononuclear samples of 4 DF and 4 DHF patients with clear epidemiological characteristics and medical history were collected. The PCR product from E region of DEN-2 isolates from blood in acute phase were ligased with the T/A cloning vector, and PCRⅡ-TOPO was used to transform to E. coli TOP 10 competent cells. About 12~20 clones were randomly selected, completely E gene sequenced and aligned, p-distance was used in analyzing sequence diversity. Our results showed that there was no relationship between DHF and more diversity of quasispecies at E region but those dengue patients had underlying diseases demonstrated higher diversity (p=0.072, Mann-Whitney test).
In summary, the Cosomopolitan genotype of DEN-2 came into Taiwan in 2001 and increased its diversity in 2002-2003 plus geographical variation taken together had facilitated to emerge more variants of DEN-2. Furthermore, the rapid infection of large human populations including patients with underlying diseases increased the dimension of variability of DEN-2. Therefore, diversified DEN-2 strains through a series of human-mosquito-human transmission chains interacting with host and environmental factors might emerge the phenotype of DEN-2 with more virulence or higher epidemic potentials. Therefore, future studies on pathogenesis of DHF need to emphasize the interactions between virus evolution and alternating hosts through transmission chains.

並列關鍵字

Dengue Fever ； Dengue Hemorrhagic Fever ； Epidemiology

參考文獻

蘇益仁 (2003). 登革熱防治工作手冊, 行政院衛生署疾病管制局.

Barnes, W. J. and L. Rosen (1974). "Fatal hemorrhagic disease and shock associated with primary dengue infection on a Pacific island." Am J Trop Med Hyg 23(3): 495-506.

Falconar, A. K. (1999). "Identification of an epitope on the dengue virus membrane (M) protein defined by cross-protective monoclonal antibodies: design of an improved epitope sequence based on common determinants present in both envelope (E and M) proteins." Arch Virol 144(12): 2313-30.

Gritsun, T. S., E. C. Holmes, et al. (1995). "Analysis of flavivirus envelope proteins reveals variable domains that reflect their antigenicity and may determine their pathogenesis." Virus Res 35(3): 307-21.

Gubler, D. J. (1987). "Dengue and dengue hemorrhagic fever in the Americas." P R Health Sci J 6(2): 107-11.