許菁芳Hsu,Jing-Fang陳怡蒨Chen,I-ChienChen國際健康產業經營管理碩士在職專班2026-03-232030/8/1020252025/7/10https://handle.ncl.edu.tw/11296/29e7rqhttps://scholars.csmu.edu.tw/handle/123456789/6466本研究針對臺灣南部某造船公司,於2015年(104年)與2024年(113年)期間針對焊接與切割等熱加工作業者進行橫斷式暴露評估,整合環境監測、個人採樣與生物監測資料,探討金屬燻煙之暴露特性、健康風險與潛在防護策略。研究結果摘要如下:104年區域環境監測顯示,電焊與切割區錳(Mn)平均濃度分別為9.51與7.45 μg/m³,雖未超過職業暴露容許濃度(Permissible Exposure Limit, PEL)10%,但明顯高於國內造船業背景值(<1 μg/m³),依美國工業衛生學會(American Industrial Hygiene Association, AIHA)分類屬「低度暴露區」,惟該分類未涵蓋微粒粒徑、暴露頻率與個體差異等健康影響要素。相對而言,個人採樣結果顯示更高暴露量,例如104年鐵(Fe)個人採樣濃度達108 μg/m³,顯著高於相對應環境濃度(焊接區44.88、切割區86.1 μg/m³);113年錳(Mn)個人採樣濃度為10.21 μg/m³,亦遠高於該年環境平均值(焊接區0.468、切割區0.271 μg/m³),顯示個人採樣應作為風險評估之優先依據。生物監測部分,113年焊接作業者於上工前後之機次比較顯示,錳(Mn)與8-OHdG(8-hydroxy-2'-deoxyguanosine)濃度皆顯著上升(錳(Mn):2.05 → 4.57 μg/L;8-OHdG:101.89 → 156.93 ng/mg creatinine),反映實際作業活動下金屬暴露與氧化壓力反應急遽升高,支持導入區域化與作業類型分層之防護策略。反觀104年橫斷面資料中,金屬暴露與8-OHdG之間尚未呈現統計顯著關聯(如錳(Mn):r = 0.005,p = 0.977),但身體質量指數(Body Mass Index, BMI)與8-OHdG具中度正相關(r = 0.299,p = 0.068),提示個人因子亦為潛在干擾來源,應納入控制變項中。整體而言,氧化性DNA損傷指標8-OHdG於113年顯著上升,由104年之3.81 ± 1.74增至86.67 ± 75.97 ng/mg creatinine,顯示勞工體內氧化壓力反應明顯惡化,需警覺其潛在慢性健康影響。雖104與113年均使用液相層析質譜儀(Liquid Chromatography–Mass Spectrometry, LC-MS)進行分析,但因由不同實驗室執行,可能產生系統性偏差,加上113年未同步蒐集吸菸、飲食與年齡等背景資料,限制差異解釋力。建議後續研究應統一實驗室流程,並納入個體問卷資料以提升分析信度。113年新增之外部辦公室對照組中,鋁(Al)與鎳(Ni)濃度(分別為1.143與0.298 μg/m³)高於造船作業區(分別為0.411與0.008 μg/m³),可能與空間設計、通風條件或鄰近污染源有關,顯示通勤動線與背景暴露不容忽視,並驗證「相似暴露群體(Similar Exposure Group, SEG)」分群方法於實務監測與風險評估之應用價值。健康風險評估方面,104年錳(Mn)、鉻(Cr)與銅(Cu)之非致癌危害指數(Hazard Quotient, HQ)分別為69.4、20.3與1.8,均超過美國環保署(U.S. Environmental Protection Agency, USEPA)建議門檻(HQ > 1);113年錳(Mn)之HQ更升至204.2,為首要風險來源。部分高風險金屬如鉻(Cr)與銅(Cu)於113年因材料更換而不再使用或濃度低於偵測極限,未納入該年風險評估。104年鉻(Cr)(推定為Cr(VI))致癌風險值(Cancer Risk, CR)為2.6×10⁻³,遠超USEPA建議容許範圍(10⁻⁶~10⁻⁴),應列為優先控制金屬。總非致癌風險指數(Non-carcinogenic Hazard Index, NHI)亦由104年之91.85上升至113年之204.2,反映風險結構雖有轉變,總體負荷並未降低。本研究強調,單點濃度監測難以全面反映實際吸入暴露與生理反應,應結合個人採樣與生物指標進行「暴露–效應」雙重監測。另應依作業區域與工項條件設計差異化防護策略,並導入可即時調整之監測制度,以因應產線變動與空間重組所引發之暴露風險變化。This cross-sectional study assessed welding and cutting-related metal fume exposure among shipyard workers in southern Taiwan during two time points: 2015 and 2024. It integrated area sampling, personal sampling, and biomonitoring data to evaluate exposure levels, oxidative stress responses, and health risk implications, aiming to inform occupational protection strategies.In 2015, area concentrations of manganese (Mn) were 9.51 μg/m³ in welding zones and 7.45 μg/m³ in cutting zones, both significantly higher than domestic shipbuilding background levels (<1 μg/m³) but below 10% of the permissible exposure limit (PEL). According to the American Industrial Hygiene Association (AIHA), these areas were categorized as “low exposure,” although such classification does not account for particle size, exposure frequency, or individual variability. Personal sampling revealed substantially higher exposure: iron (Fe) in 2015 reached 108 μg/m³, and Mn in 2024 reached 10.21 μg/m³, both exceeding area averages, indicating that personal exposure is more representative for risk estimation and control planning.Biomonitoring showed acute exposure responses in 2024: among welders, Mn and 8-hydroxy-2'-deoxyguanosine (8-OHdG) concentrations rose significantly after work shifts (Mn: 2.05 → 4.57 μg/L; 8-OHdG: 101.89 → 156.93 ng/mg creatinine), supporting area-specific control measures based on task type. In contrast, 2015 cross-sectional data did not reveal statistically significant correlations between urinary metals and 8-OHdG levels (e.g., Mn: r = 0.005, p = 0.977), although BMI showed a moderate positive trend with 8-OHdG (r = 0.299, p = 0.068). These findings suggest that personal factors may confound exposure-effect relationships and should be included in future study designs.The average urinary 8-OHdG concentration increased sharply from 3.81 ± 1.74 ng/mg creatinine in 2015 to 86.67 ± 75.97 in 2024, indicating elevated oxidative stress. Although both datasets were analyzed using LC-MS (Liquid Chromatography–Mass Spectrometry), different laboratories were involved, and individual background factors (e.g., smoking, diet) were not collected in 2024, limiting explanatory power. Standardizing analytical procedures and integrating questionnaire-based background data are recommended.In 2024, a non-shipbuilding office control group exhibited higher aluminum and nickel concentrations (Al: 1.143 μg/m³; Ni: 0.298 μg/m³) than the shipyard exposure group, likely due to ventilation conditions or nearby pollution sources. This confirms the importance of Similar Exposure Group (SEG) stratification in exposure assessment and practical risk management.Health risk assessments identified Mn as the dominant risk metal, with HQs of 69.4 in 2015 and 204.2 in 2024. Cr and Cu also exceeded acceptable HQ levels in 2015 (20.3 and 1.8, respectively). Notably, hexavalent chromium presented a cancer risk of 2.6 × 10⁻³, well above the US EPA guideline (10⁻⁶–10⁻⁴). Although Cr and Cu were no longer in use or below detection in 2024, overall non-carcinogenic hazard indices (NHI) increased from 91.85 to 204.2, indicating that cumulative exposure risks remain unresolved despite reduced metal types.This study underscores the limitations of single-time-point area sampling for health risk assessment and recommends dual monitoring through personal exposure and biological effect indicators. In addition, production restructuring in 2024 altered task distributions and exposure conditions, highlighting the need for dynamic, task-specific monitoring systems and adaptive risk communication in occupational health programs.Special thanks to Professor Hung-Hsin Liu for his valuable guidance.造船業熱加工作業重金屬燻煙個人採樣8-OHdG生物監測氧化壓力非致癌風險致癌風險相似暴露群體shipbuilding industrythermal processing operationsheavy metal fumespersonal samplingbiomonitoringoxidative stressnon-carcinogenic riskcarcinogenic riskSEG造船業重金屬暴露調查與健康風險評估thesis