摘要目的:评估Kawasaki法、INTERSALT法和Tanaka法估算中国人群24 h尿钠排泄量的可行性。方法:采用多阶段分层随机抽样方法，于2017年在浙江省抽取义乌市、海宁市、泰顺县、宁波市鄞州区、丽水市莲都区的1 499名18~69岁居民，采用问卷调查收集基本信息并进行体格测量，收集24 h尿液后记录尿量并检测尿钠、尿钾和尿肌酐浓度，最终共有1 426名调查对象通过了尿液完整性检测。分别应用Kawasaki法、INTERSALT法和Tanaka法估算24 h尿钠排泄量，并与实测值进行比较。结果:调查对象年龄为（46.71±14.04）岁，其中男性700名，占49.1%。调查对象24 h尿钠排泄量实测值为（167.10±74.70）mmol，Kawasaki法估值偏高，为（184.61±57.10）mmol，INTERSALT法和Tanaka法估值偏低，分别为（134.62±39.21）和（143.20±35.66）mmol。估算差值（95 %CI）（mmol）由小及大依次为Kawasaki法[17.51（13.54，21.47）]、Tanaka法[-23.90（-27.60，-20.20）]和INTERSALT法[-32.48（-36.29，-28.67）]。随着24 h钠盐摄入量的增加，各估算方法均由高估转为低估：在24 h钠盐摄入量<9.0 g者中，INTERSALT法估算差值（95 %CI）最小，在<6.0和6.0~8.9 g者中分别为[43.15（37.73，48.57）]和[1.26（-2.10，4.63）]mmol；在24 h钠盐摄入量≥9.0 g者中，Kawasaki法估算差值（95 %CI）最小，在9.0~11.9和≥12.0 g者中分别为[-12.50（-18.14，-6.86）]和[-53.73（-61.25，-46.22）]。Bland-Altman法一致性分析显示，Kawasaki法估算值与实测值的一致性最好，范围外点数最少（69个，占4.84%）。 结论:在3种点尿法中，Kawasaki法对预测中国人群24 h尿钠排泄量具有更好的适用性。

abstractsObjective:To evaluate the feasibility of three spot urine methods (Kawasaki, INTERSALT and Tanaka) for estimating the 24 h urinary sodium excretion in the Chinese population.Methods:In 2017, 1 499 participants aged 18 to 69 years old were selected from Yiwu City, Haining City, Taishun County, Yinzhou District of Ningbo City and Liandu District of Lishui City of Zhejiang Province by using the multistage random sampling method. Sociodemographic information of the subjects was collected with questionnaires and physical measurements were performed. 24 h urine was collected and urinary volume was recorded. The concentrations of urinary sodium, potassium and creatinine were also measured. Kawasaki, INTERSALT and Tanaka spot urine methods were applied to estimate the 24 h urinary sodium excretion and compared with actual values among 1 426 participants who passed urine integrity test.Results:The age of participants was (46.71±14.04) years old, including 700 males, accounting for 49.1%. The actual value of 24 h urinary sodium excretion was (167.10±74.70) mmol, but Kawasaki method overestimated it as (184.61±57.10) mmol, and INTERSALT and Tanaka methods underestimated it as(134.62±39.21) and (143.20±35.66) mmol. Estimated difference (95 %CI) (mmol) from small to large was Kawasaki method [17.51 (13.54, 21.47)], Tanaka method [-23.90 (-27.60, -20.20)] and INTERSALT method [-32.48 (-36.29, -28.67)]. With the increase of 24 h sodium intake, all estimation methods changed from the overestimation to underestimation. In those with 24 h sodium intake <9.0 g, the estimated difference (95 %CI) of the INTERSALT method was the smallest as 43.15 (37.73, 48.57) and 1.26 (-2.10, 4.63) mmol for <6.0 and 6.0-8.9 g groups, respectively. In those with 24 h sodium intake≥9.0 g, the Kawasaki method had the smallest estimated difference (95 %CI) as -12.50 (-18.14, -6.86) and -53.73 (-61.25, -46.22) for 9.0-11.9 g and ≥ 12.0 g group, respectively. The consistency analysis of the Bland-Altman method showed that the Kawasaki method had the best consistency with actual measured value and it had the least number of points outside the range (69 points accounting for 4.84%). Conclusion:Among the three spot urine methods, the Kawasaki method has better applicability in predicting the excretion of 24 h urine sodium in the Chinese population.