目的 上海市疾病预防控制中心（简称上海市疾控）对各区疾控送检的疟疾病例血样进行复核,并对复核结果进行分析,为进一步提升各区疾控疟疾诊断水平提供依据。方法 2017—2019年上海市疾控通过镜检和疟原虫核酸检测（巢式PCR）分别对各区疾控疟疾实验室送检的血涂片和血样进行复核。以市疾控结果为标准,分析各区疾控疟疾实验室镜检和疟原虫核酸检测的符合率,以及核酸检测敏感性和特异性。并于2019年对各区疾控进行了一次疟原虫核酸检测盲样考核。结果 2017—2019年区疾控共送检疟疾复核病例样品232份,其中完整的样品225份（同时具备血涂片和血样）,完整率99.1%（225/227）。225份完整的样品中,金山区占44.0%（99/225）,静安区占15.6%（35/225）,浦东新区占7.6%（17/225）,虹口区占6.7%（15/225）,青浦区和崇明区较少,分别为0份和1份,其余各区送样量均占比6.0%以下。225份样品中,区疾控与市疾控的镜检总符合率、阳性符合率和阴性符合率分别为96.0%（216/225）、98.8%(170/172)和86.8%(46/53),疟原虫核酸检测的符合率为88.8%（103/116）。区疾控疟疾核酸检测采用实时荧光PCR（qPCR）法,敏感性为76.6%（85/111）,与市疾控巢式PCR法87.4%（97/111）的敏感性差异有统计学意义（P < 0.05）;市区两级疾控的两种核酸检测方法的特异性均为100%。上述2种检测方法对225份样品进行综合评价,区疾控的符合率为96.9%（218/225）;误判率3.1%（7/225）,其中定性错误5份,定种错误2份。2019年,市疾控对区疾控进行的疟疾核酸盲样考核的总正确率为97.5%（78/80）,其中恶性疟原虫样品的正确率为94.1%（32/34）,与间日疟、卵形疟、三日疟原虫和阴性样品的正确率差异均无统计学意义（ P > 0.05）。 结论 2017—2019年上海市各区疾控疟疾实验室的镜检能力较强,但疟原虫核酸检测的敏感性尚待提高。

Objective To recheck the blood samples of malaria cases sent by district-level centers for disease control and prevention (CDC), in Shanghai Municipal Center for Disease Control and Prevention (SCDC), and analyze the review results, in order to provide a basis for improving the malaria diagnostic capability of district CDCs.Methods Blood smears and whole blood samples submitted by the district CDC malaria laboratories were rechecked by SCDC using microscopic examination and nuclear acid detection (nested PCR) during 2017—2019. Using the examination by SCDC as the standard, the positive and negative coincidence rates of microscopic examination, and the sensitivity and specificity of malaria nucleic acid detection performed in district CDC malaria laboratories were analyzed. In 2019, blind assessment of malaria nucleic acid tests was conducted for all district CDCs of each district.Results A total of 232 malaria samples were submitted by district CDCs for rechecking in 2017—2019, among them 225 were complete samples (containing both blood smear and whole blood samples), with a completeness rate of 99.1% (225/227). Of the 225 complete samples, 99 samples were from Jinshan District (44.0%); 35 from Jing’an District (15.6%); 17 from Pudong New District (7.6%); 15 from Hongkou District (6.7%); Qingpu District and Chongming District submitted the least numbers of samples (0 and 1, respectively); while other districts provided less than 6.0% of the samples. The recheck of 225 complete samples showed an overall coincidence rate, positive coincidence rate and negative coincidence rate being 96.0% (216/225), 98.8% (170/172) and 86.8% (46/53), respectively, and the coincidence rate of nucleic acid detection 88.8% (103/116). The district CDCs used the quantitative real-time PCR (qPCR) method for nucleic acid detection, resulting in a sensitivity of 76.6% (85/111), which was significantly different from that of the nested PCR method used by SCDC (87.4%, 97/111) (P < 0.05). The specificity of the two nucleic acid detection methods was 100.0%. Comprehensive evaluation of the 225 samples by using the two detection methods showed that the coincidence rate of district CDCs was 96.9% (218/225), and the misjudgment rate was 3.1% (7/225), including 5 with qualitative errors and 2 with errors in species identification. In 2019, the overall correct detection rate of blind assessment of malaria nucleic acid by district CDCs was 97.5% (78/80). Specifically, the correct detection rate of P. falciparum sample was 94.1% (32/34), which was not significantly different from those of the other four types of samples (P. vivax, P.ovale, P. malariae and negative infection) (P > 0.05). Conclusion From 2017 to 2019, the malaria laboratories of district CDCs showed a strong capability of malaria microscopic examination, but the sensitivity of nuclear acid detection needs to be improved.