【摘要】 目的 探討CT對嬰幼兒腹部巨大囊性病變的診斷價值及其鑒別診斷。 方法 搜集2003年-2009年間經手術病理證實的2歲以內患兒腹部巨大囊性病變62例,分析其病變部位、大小、形態、分隔、密度和強化、囊壁及與周圍臟器關系等要素。 結果 62例中,先天性膽總管囊腫4例,重度腎積水36例,巨輸尿管2例,囊性腎母細胞瘤4例,大網膜囊腫4例,腸系膜囊腫3例,卵巢囊腫6例,囊性畸胎瘤3例。 各種囊性病變有其一定的發病部位和特征性的CT表現。 結論 CT是嬰幼兒腹部囊性病變定位、定性診斷的重要影像學方法。【Abstract】 Objective To explore the value of CT diagnosis and differential diagnosis of the giant cystic lesions in abdomen of the infants. Methods A total of 62 infants younger than 2 years old with the giant cystic lesions in abdomen confirmed by surgery and histopathology from 2003 to 2009 were collected. The location of the lesion, range, configuration, thickness of cystic wall and septa, density, contrast enhancement, and adjacent organs were observed and analyzed. Results In 62 infants, there were congenital cyst of common bile duct in 4, giant hydronephrosis in 36, primary megaureter in 2, cystic Wilms tumor in 4, greater omentum cyst in 4, mesenteric cyst in 3, ovarian cystic in 6, and cystic teratoma in 3. Each disease had its own lesions location and features of CT images. Conclusion CT is very effective on the localized and qualitative diagnosis of the giant cystic lesions in abdomen of infants.
目的介紹利用 SAS MIXED 和 SAS NLMIXED 實現線性或非線性多水平模型的 Meta 分析。方法以 Shim 等發表的輔助手術降低宮頸癌根治性同步放化療患者局部復發風險的系統評價結果作為實例數據,利用 SAS 軟件實現基于線性或非線性多水平模型的 Meta 分析,并提供編程代碼。結果在沒有協變量情況下,基于雙變量隨機效應模型的 PROC MIXED 和非線性混合效應模型的 PROC NLMIXED 的 OR 合并效應值分別為[0.63,95%CI(0.46,0.87),P=0.005 7]和[0.60,95%CI(0.39,0.81),P=0.000 3]。在帶有協變量情況下,雙變量隨機效應模型和非線性混合效應模型 OR 效應值為[0.65,95%CI(0.47,0.91),P=0.011]和[0.59,95%CI(0.38,0.80),P=0.000 3]。協變量 OR 效應值分別為[2.70,95%CI(0.16,45.23),P>0.05]和[1.86,95%CI(?0.07,3.79),P=0.06]。結論利用 SAS NLMIXED 非線性混合效應模型擬合的 Meta 分析結果與 SAS MIXED 線性混合效應模型的 Meta 分析結果相似,鑒于 PROC NLMIXED 具有強大的編程能力及非線性混合效應模型對稀疏數據具有靈活的建模能力,PROC NLMIXED 在 Meta 分析領域將發揮越來越重要的作用。