A Diagnostic Tool for Identification of Etiologies of Fever of Unknown Origin in Adult Patients

The diagnosis and treatment of fever of unknown origin (FUO) are huge challenges to clinicians.Separating the etiologies of FUO into infectious and non-infectious disease is conducive to clinical physicians not only on making decisions rapidly concerning the prescription of suitable antibiotics but also on further analysis of the final diagnosis.In order to develop and validate a diagnostic tool to efficiently distinguish the etiologies of adult FUO patients as infectious or non-infectious disease,FUO patients from the departments of infectious disease and internal medicine in three Chinese tertiary hospitals were enrolled retrospectively and prospectively.By using polynomial logistic regression analysis,the diagnostic formula and the associated scoring system were developed.The variables included in this diagnostic formula were from clinical evaluations and common laboratory examinations.The proposed tool could discriminate infectious and noninfectious causes of FUO with an area under receiver operating characteristic curve (AUC) of 0.83,sensitivity of 0.80 and specificity of 0.75.This diagnosis tool could predict the infectious and non-infectious causes of FUO in the validation cohort with an AUC of 0.79,sensitivity of 0.79 and specificity of 0.70.The results suggested that this diagnostic tool could be a reliable tool to discriminate between infectious and non-infectious causes of FUO.

Establishment of a Predictive Model Related to Pathogen Invasion for Infectious Diseases and Its Diagnostic Value in Fever of Unknown Origin

The present study aimed to establish a list of parameters indicative of pathogen invasion and develop a predictive model to distinguish the etiologies of fever of unknown origin (FUO) into infectious and non-infectious causes.From January 2014 to September 2017,431 patients with FUO were prospectively enrolled in the study population.This study established a list of 26 variables from the following 4aspects:host factors,epidemiological factors,behavioral factors,and iatrogenic factors.Predefined predicted variables were included in a multivariate logistic regression analysis to develop a predictive model.The predictive model and the corresponding scoring system were developed using data from the confirmed diagnoses and 9 variables were eventually identified.These factors were incorporated into the predictive model.This model discriminated between infectious and non-infectious causes of FUO with an AUC of 0.72,sensitivity of 0.71, and specificity of 0.63.The predictive model and corresponding scoring system based on factors concerning pathogen invasion appear to be reliable screening tools to discriminate between infectious and non-infectious causes of FUO.

Case Report: Fever of Unknown Origin <br/>—An Unusual Presentation for Diffuse Large B-Cell Lymphoma

A 61-year-old male initially presented with fever of unknown origin. He had extensive work-up over two years including an infectious diseases panel, autoimmune studies, and Rheumatology and Hematology evaluations. The patient was initially diagnosed with Adult Still’s disease and underwent an out-patient right nodal fine-needle aspiration that was indeterminate. After continued failure of treatment for Adult Still’s disease, the patient had surgical resection of a right axillary lymph node that yielded the diagnosis of diffuse large B-cell lymphoma. Further work-up revealed Epstein-Barr virus positivity, the possible trigger behind his mutation for diffuse large B-cell lymphoma and its uncommon presentation. The patient met criteria for central nervous system prophylaxis and received multiple administrations throughout his therapy. He ultimately expired following recurrence of his disease at its initial site but without central nervous system involvement. We report an uncommon presentation of a patient with diffuse large B-cell lymphoma. This lymphoma can have numerous, vague presentations requiring a broad differential diagnosis and may lead to multiple evaluations prior to an ultimate diagnosis. We will also discuss the need for central nervous system prophylaxis, how this patient is qualified for prophylaxis, and how central nervous system prophylaxis benefits, harms, or does not affect patients with diffuse large B-cell lymphoma.

Clinical Features of Spontaneous Remission in the Classic Fever of Unknown Origin:A Retrospective Study

Objective To summarize the clinical features of spontaneous remission in classic fever of unknown origin(FUO).Methods Medical records of 121 patients diagnosed with FUO at admission in Peking Union Medical College Hospital between January 2018 and June 2018 were reviewed retrospectively.Patients who were discharged without etiological diagnoses were followed for 2 years.The clinical features and outcomes of these patients were summarized.Multivariate logistic regression was used to analyze related factors of spontaneous remission of FUO.Results After excluding 2 patients who lost to follow-up,the etiology of 119 FUO patients were as follows:infectious diseases in 30(25.2%)cases,connective tissue diseases in 28(23.5%)cases,tumor diseases in 8(6.7%)cases,other diseases in 6(5.0%)cases,and unknown diagnoses in 47(39.5%)cases.Totally,41 patients experienced spontaneous remission of fever(the median time from onset to remission was 9 weeks,ranging from 4 to 39 weeks).In patients with spontaneous remission in FUO,lymphadenopathy was less common clinical manifestation,the levels of inflammatory markers including leukocyte count,neutrophil count,neutrophil ratio,C-reactive protein,and ferritin were lower,and the proportion of CD8 positive T lymphocytes expressing CD38 was lower.Multivariate logistic regression analysis of factors with a P-value<0.05 in univariate analysis shown that white blood cell count(OR:0.S45,95%CI:0.306-0.971,P=0.039),neutrophil count(OR:2.074,95%CI:1.004-4.284,P=0.049),and proportion of neutrophils(OR:0.928,95%Cl:0.871-0.990,P=0.022)were independent significant factors associated with spontaneous remission in FUO.Conclusions This study suggested that most patients discharged with undiagnosed classic FUO would remit spontaneously.Thus,for patients with stable clinical conditions,follow-up and observation could be the best choice.Patients with lower level of some inflammatory factors may have a high likelihood of spontaneous remission in classic FUO.

Large renal carbuncle mimicking intra-abdominal neoplasm on Ga-67 scintigraphy: A case of fever of unknown origin

Fever of unknown origin is still a medical challenge. 67Ga single photon emission computed tomography/computed tomography images are commonly used to evaluate the final diagnosis of fever of unknown origin. We presented a case of fever of unknown origin undergone 67Ga scintigraphy and an intra-abdominal 67Ga avid tumor was detected which suspected to be a neoplasm. Further contrast enhanced computed tomography revealed that the lesion was a large renal carbuncle. We concluded that the contrast-enhanced CT or single photon emission computed tomography/computed tomography with contrast enhancement can be performed to further improve diagnostic performance.

Infectious causes of fever of unknown origin in developing countries: An international ID-IRI study

Background:Fever of unknown origin(FUO)in developing countries is an important dilemma and further research is needed to elucidate the infectious causes of FUO.Methods:A multi-center study for infectious causes of FUO in lower middle-income countries(LMIC)and lowincome countries(LIC)was conducted between January 1,2018 and January 1,2023.In total,15 participating centers from seven different countries provided the data,which were collected through the Infectious DiseasesInternational Research Initiative platform.Only adult patients with confirmed infection as the cause of FUO were included in the study.The severity parameters were quick Sequential Organ Failure Assessment(qSOFA)≥2,intensive care unit(ICU)admission,vasopressor use,and invasive mechanical ventilation(IMV).Results:A total of 160 patients with infectious FUO were included in the study.Overall,148(92.5%)patients had community-acquired infections and 12(7.5%)had hospital-acquired infections.The most common infectious syndromes were tuberculosis(TB)(n=27,16.9%),infective endocarditis(n=25,15.6%),malaria(n=21,13.1%),brucellosis(n=15,9.4%),and typhoid fever(n=9,5.6%).Plasmodium falciparum,Mycobacterium tuberculosis,Brucellae,Staphylococcus aureus,Salmonella typhi,and Rickettsiae were the leading infectious agents in this study.A total of 56(35.0%)cases had invasive procedures for diagnosis.The mean qSOFA score was 0.76±0.94{median(interquartile range[IQR]):0(0–1)}.ICU admission(n=26,16.2%),vasopressor use(n=14,8.8%),and IMV(n=10,6.3%)were not rare.Overall,38(23.8%)patients had at least one of the severity parameters.The mortality rate was 15(9.4%),and the mortality was attributable to the infection causing FUO in 12(7.5%)patients.Conclusions:In LMIC and LIC,tuberculosis and cardiac infections were the most severe and the leading infections causing FUO.

基于双随机森林的发热待查智能诊断方法

在机器学习预测模型中,不平衡数据集会降低少数类的预测准确性。针对发热待查数据集的不平衡特性,该文提出了一种基于K-Means聚类欠采样的双随机森林病因预测方法。首先通过K-Means聚类欠采样构建一个平衡数据集,并在此基础上创建一个基于CART投票机制的随机森林预测模型。然后对初始数据集用同样的方法创建一个随机森林预测模型。最后将两个随机森林预测模型联合,使用两者的CART一起投票预测。该文提出的方法增加了CART的数量,在保持原有数据集特性的同时,提高了少数类的投票权重。在发热待查数据集上的实验表明,该文所提方法不仅改善了少数类的预测性能,对其他类别的预测性能也有一定程度的提升。

从热邪、湿邪角度探讨不明原因发热的中医治疗

不明原因发热病因不明,西医治疗往往困难。总结了10余年诊治经验,发现采用中医治疗,疗效明显。不明原因发热属于中医内伤发热范畴,患者常有气、血、阴、阳亏虚或气郁、血瘀、湿阻病史,导致五邪产生,发热为内生邪气所致,病理性质属热。患者多苔厚黏腻,发热反复,从舌苔表现及病程缠绵判断,病理性质属于湿邪。因无外感诱因,故其湿邪属于内湿。因此,不明原因发热患者的临床表现主要体现在热邪、湿邪两方面,属于湿热并重。因病程久,容易阻滞三焦气机,日久湿、热、毒邪壅滞,弥漫三焦。治疗原则宜清上、畅中、渗下同用,清热、利湿、解毒并行,需较强力度用药方可使弥漫三焦之湿热毒邪俱除。遣方用药上,甘露消毒丹用药作用于上下、内外,升降出入,莫不关乎三焦之枢,故为最佳之选。采用力度较轻的王氏连朴饮、达原饮、杏仁滑石汤、三仁汤治疗,效果不如甘露消毒丹。研究从发热的特点、治疗思路、治法确立、方剂选择等方面阐述了不明原因发热的中医治疗,以期为临床提供参考。

以伪装热为表现的儿童孟乔森综合征10例病例系列报告

背景以伪装热为表现的儿童孟乔森综合征病例常就诊于精神科、心理科以外的其他科室,这些科室医生对该病认识不足,患儿很难得到及时和正确的诊治,发病率可能被低估。目的 总结以伪装热为表现的孟乔森综合征儿童的临床资料,提高对该病的认识。设计病例系列报告。方法 回顾性纳入2022年1月至2024年1月在3个医院诊断以伪装热为表现的孟乔森综合征的连续住院病例。通过各医院管理信息系统采集病例性别、发病年龄、病程,就诊经历及治疗史,发热特征,体格检查,辅助检查等临床资料。主要结局指标发热特征。结果 10例患儿进入本文分析,均为女童,起病年龄12(10~14)岁,病程6.3(1~18)个月。门诊频次平均8(4~12)次,住院频次平均3.5(2~5)次,就诊医院数平均3.5(3~5)家。均有反复行糖皮质激素、抗菌药物治疗史且均无疗效。均在短时间内体温达到高热(39.0~40.2℃),约5 min内体温恢复正常;6例偶有头晕,其中伴头痛2例,伴腹痛1例;发热前均无畏寒及寒战,时间无规律性;体温恢复正常时无出汗反应。发热前后皮肤体感温度、脉搏、心率、呼吸频次均未见变化。精神和食欲均正常;体重轻度增加3例,余病例均未见下降;均无阳性体征。辅助检查均未见异常。单亲家庭环境4例,父母长期争执、吵闹5例,留守儿童1例。均转至心理科进行干预,随访时间8(5~12)个月,未见类似发热发作,也无因其他症状就诊者。结论 以伪装热为表现的儿童孟乔森综合征好发于女性儿童/青少年,多次反复求治于不同医院,体温可在3~8 min内达到高热,约5 min内恢复正常,发作时无发热时典型病理生理过程变化,实验室检验和辅助检查结果无异常;虽然病程较长,但无疾病状态改变。

发热待查183例病因分析

目的了解183例发热待查(fever of unknown origin,FUO)患者的病因分布特点,比较其在不同医院、年龄段、性别、热程的病因构成,为临床诊治提供参考。方法回顾性分析郑州市第六人民医院及北京大学第一医院感染疾病科2022年8月至2023年10月收治的183例FUO住院患者的临床资料,按病因分为感染性疾病93例、非感染性炎症性疾病(non-infectious inflammatory diseases,NIID)42例、肿瘤性疾病14例、其他疾病21例、诊断不明13例,比较在不同医院、不同年龄段、不同性别、不同热程的病因分布差异。采用χ^(2)检验进行比较。结果纳入FUO患者中男性占61.20%(112/183),年龄(49.70±19.09)岁,细菌感染、成人Still病、淋巴瘤、坏死性淋巴结炎分别为各组确诊患者中常见病因。两家医院FUO患者中感染性疾病例数占比分别为64.29%(27/42)、46.81%(66/141),差异有统计学意义(χ^(2)=3.96,P<0.05);31~45岁患者感染性疾病数占比高于15~30岁患者[63.89%(23/36)比34.15%(14/41),χ^(2)=6.79,P<0.05];15~30岁患者其他疾病占比[29.27%(12/41)]高于其他各组患者(χ^(2)=4.10、6.35、9.51,均P<0.05);女性NIID占比高于男性[33.80%(24/71)比16.07%(18/112),χ^(2)=7.73,P<0.05];热程>3个月的感染性疾病患者占比[12.50%(3/24)]低于热程<1个月与热程1~3个月患者[61.33%(46/75)、52.38%(44/84)](χ^(2)=15.45、10.51,均P<0.05);热程>3个月的肿瘤性疾病患者占比高于热程<1个月的患者[20.83%(5/24)比4.00%(3/75);χ^(2)=4.86,P<0.05];热程>3个月的诊断不明病例数占比高于热程<1个月的患者[16.67%(4/24)比2.67%(2/75),χ^(2)=4.04,P<0.05]。结论对于FUO患者,应首先考虑感染性疾病。不同医院、年龄、性别、热程的FUO患者病因构成可能有所不同,应结合患者临床特点进行精确、快速的相关检验及检查以尽快明确病因。

李创鹏从“温少阴,通阳明”论治不明原因发热

不明原因发热是临床诊疗中的难点,患者常以发热为主诉就诊,伴随症状和体征一般不典型,现代医学对于不明原因发热发病机制的研究尚无统一定论,治疗仍以经验性抗炎为主,存在诊断延迟、治疗方案有限、治疗效果欠佳、不良反应明显等不足。中医认为,不明原因发热总属脏腑功能失调,阴阳失衡,具有病程长、病因不明、病情复杂、反复发作、缠绵难愈的特征。李创鹏主任医师基于六经辨证体系,指出不明原因发热的病机特点为少阴阳明合病,提出“温少阴,通阳明”的论治观点,即运用大黄附子汤合薏苡附子败酱散退热祛邪,以细辛引药直达少阴病所,同时健运脾胃、养阴生津、通调血脉,使邪去正安。

宏基因组二代测序协助诊断重症立克次体感染一例并文献复习

立克次体感染是一种急性传染性疾病,多为自然疫源性。由于立克次体感染早期多表现为非特异性症状,如发热、皮疹、疼痛(多为四肢肌肉痛及头痛)等,且传统检验方法难以明确病原体,导致诊断困难,极易漏诊及误诊。本例患者以消化道症状起病,常规抗感染治疗效果不明显且病情迅速进展,出现多器官功能衰竭,后通过宏基因组二代测序(Metagenomic next-generation sequencing,mNGS)于外周血中检出日本立克次体病原体,并进行针对性治疗及支持治疗,病情好转出院。

Hemophagocytic lymphohistiocytosis triggered by relapsing polychondritis:A case report

BACKGROUND Hemophagocytic lymphohistiocytosis(HLH)is a rare,life-threatening disorder caused by abnormal histiocytes and T cell activation.In adults,it is predominantly associated with infections,cancers,and autoimmune diseases.Relapsing polychondritis(RP),another rare disease,is diagnosed based on symptoms without specific tests,featuring cartilage inflammation characterized by swelling,redness,and pain,rarely inducing HLH.CASE SUMMARY A 74-year-old woman visited the emergency room with a fever of 38.6℃.Blood tests,cultures,and imaging were performed to evaluate fever.Results showed increased fluorescent antinuclear antibody levels and mild cytopenia,with no other specific findings.Imaging revealed lymph node enlargement was observed;however,biopsy results were inconclusive.Upon re-evaluation of the physical exam,inflammatory signs suggestive of RP were observed in the ears and nose,prompting a tissue biopsy for confirmation.Simultaneously,persistent fever accompanied by cytopenia prompted a bone marrow examination,revealing hemophagocytic cells.After finding no significant results in blood culture,viral markers,and tissue examination of enlarged lymph nodes,HLH was diagnosed by RP.Treatment involved methylprednisolone followed by azathioprine.After two months,bone marrow examination confirmed resolution of hemophagocytosis,with normalization of hyperferritinemia and pancytopenia.CONCLUSION Thorough physical examination enabled diagnosis and treatment of HLH trig gered by RP in patients presenting with fever of unknown origin.

成人斯蒂尔病与其他不明原因发热疾病的鉴别诊断指标

目的探讨成人斯蒂尔病(AOSD)与其他不明原因发热(FUO)疾病的鉴别诊断指标。方法收集2010年1月-2021年5月在陆军军医大学第一附属医院住院的177例AOSD患者及163例待鉴别FUO患者的临床资料及实验室指标,随机分为训练组及验证组。通过单因素分析提取有统计学意义的变量进行受试者工作特征(ROC)曲线分析并获取变量的最佳截断值,进一步通过多因素logistic回归分析筛选出具有鉴别诊断意义的指标,并构建列线图模型;采用ROC曲线、校准曲线及决策曲线分析列线图的准确性及稳定性。结果单因素分析结果显示,4项临床特征(关节痛、皮疹、咽痛、肌痛)及14项实验室参数[白细胞计数(WBC)、单核细胞百分比、中性粒细胞百分比、淋巴细胞百分比、血小板计数、C反应蛋白、白细胞介素-6(IL-6)、铁蛋白、球蛋白、免疫球蛋白A、免疫球蛋白G(IgG)、肌酸激酶、肌酐、补体C3]差异均有统计学意义(P<0.05)。多因素分析结果显示,关节痛、WBC≥9.995×109/L、IL-6≥98.13ng/L、铁蛋白≥507.37ng/ml、球蛋白≤36.58g/L、IgG≤13.59g/L、补体C3≥1.27g/L均与AOSD相关(P<0.05)。训练组及验证组的曲线下面积(AUC)分别为0.917[95%可信区间(95%CI)0.883~0.951]、0.869(95%CI0.802~0.936);校准曲线表现出良好的一致性;决策曲线分析表明,训练组及验证组分别在5%~85%、10%~85%广大风险范围内显示出较大的正向收益率。结论该研究建立起一个相对准确的AOSD鉴别诊断模型,关节痛、WBC、IL-6、铁蛋白、球蛋白、IgG及补体C3多指标联合应用有助于鉴别AOSD与其他FUO病因。

肝脏及淋巴结播散性结核1例并文献复习

目的了解肝脏及淋巴结播散性结核的临床特征。方法对复旦大学附属华山医院感染科诊断的1例肝脏及淋巴结播散性结核患者的临床表现、实验室及影像学检查结果、治疗及预后进行分析,并回顾2003-2022年PubMed及中文数据库关于肝脏结核及播散性结核的文献报道,筛选并总结分析播散至肝脏和其他部位结核患者的相关临床资料。结果该例患者为老年男性,主要以反复发热、乏力纳差为主要症状,各种抗感染治疗无效,影像学检查示全身多发异常代谢增高灶,最终依赖活检明确诊断,给予抗结核治疗后,病情好转。结合相关文献和该病例,共82例肝结核合并其他部位播散患者,平均年龄39.3岁(1~59岁),多以发热起病,通过影像学检查及病灶脓液培养或活检明确结核感染,抗痨治疗后预后良好。结论肝脏及淋巴结播散结核为少见肺外结核,临床症状并无特异性。确诊依赖影像学及病理活检。早期诊断很重要。该患者确诊后使用合适的抗结核药物治疗,疗效良好。

1例罕见的限制性马拉色菌致血流感染患者的临床诊疗分析

目的:分析1例罕见的限制性马拉色菌致血流感染患者的临床诊疗过程,为临床此类罕见感染疾病的诊治提供参考。方法与结果:该患者2周前被猫抓伤后反复发热,入院后仍持续发热,检查显示血清白细胞计数和降钙素原均正常,仅C反应蛋白升高,结合患者的临床症状,临床排除了呼吸道和尿路的感染,且未发现其他明显的感染灶,临床先经验性予阿奇霉素;2 d后,真菌D-葡聚糖检查结果异常(222.53 pg/mL);临床药师会诊发现,患者近期存在糖尿病血糖控制不佳且有服用糖皮质激素的情况,因而可能免疫功能较差,综合患者前期检查结果,怀疑患者的发热可能是由猫抓伤致真菌侵入血液所致的血流感染引起的,遂建议加用氟康唑进行抗真菌治疗;第2天,患者体温明显回落,且第3天患者血标本的二代测序结果为限制性马拉色菌(序列数为29),基本印证了临床药师的判定;最后,患者在用氟康唑治疗10 d后出院。结论:对于不明原因发热的患者,临床医生和药师应充分结合患者的临床表现和各种检查结果,以及经验性抗感染治疗的反馈,以准确判断患者的感染部位和可能病原菌,从而更有针对性、更有效地为患者开展治疗。

Applied Value of Monitoring Serum Hepcidin in Differential Diagnosis of Infection versus Tumor Fevers

The applied value of serum hepcidin in differential diagnosis of infection fevers versus tumor fevers was explored.A total of 432 fever patients were selected according to the domestic fever of unknown origin（FUO） diagnostic criteria from our hospital between June 2010 and November 2013.Venous blood samples were taken on the day 1,5,10 after admission.The infection group（98 cases） and the tumor group（50 cases） were set up based on the clinical and laboratory findings.ELISA was used to determine the serum hepcidin and IL-6 levels.SPSS 13.0 was used for statistical analysis.Hepcidin showed obvious descending trend on the 10 th day in both the bacterial infection group（66 cases） and the virus infection group（32 cases）,and the descending trend was similar to that of inflammatory indexes such as procalcitonin（PCT）,hypersensitive C-reactive protein（h-CRP）,erythrocyte sedimentation rate（ESR）,white blood cell（WBC）,and ferritin.Serum hepcidin showed no obvious differences in the tumor group on the day 1,5,10 after admission.In the infection groups,serum hepcidin was positively correlated with IL-6（r=0.687,P=0.000） and CRP（r=0.487,P=0.026）,but had a poor correlation with blood sedimentation,ferritin,PCT and WBC（P〉0.05）.Monitoring dynamic changes of hepcidin and related inflammatory factors in patients with fever is expected to be used for clinical identification of infection fever and tumor fever.

红串红球菌感染1例报告并文献复习

红串红球菌是低毒力的条件致病菌,人类感染罕见,国内尚缺乏报道。本文报告1例在糖尿病基础上以发热、肺结节为主要临床表现的病例,经肺穿刺活检及宏基因组二代测序证实为红串红球菌感染,在此基础上系统回顾国内外相关文献,总结该菌感染的临床表现和诊治经验供国内外同行参考,以期今后对该菌感染能及时正确的诊断和治疗。

^(18)F-FDG PET/CT对非肿瘤性疾病导致不明原因发热的诊断价值

目的 评价^(18)F-FDG PET/CT对非肿瘤性疾病导致的不明原因发热(FUO)的诊断价值。资料与方法 回顾性分析东南大学附属中大医院2017年9月—2021年12月以FUO就诊的79例患者的^(18)F-FDG PET/CT检查结果,患者均基于临床诊断等做出最终诊断,评价^(18)F-FDG PET/CT对感染性疾病及风湿免疫性疾病的诊断贡献。结果 79例FUO患者最终确诊71例(89.9%),其中感染性疾病47例(59.5%),风湿免疫性疾病21例(26.6%),结节病1例(1.3%),药物热1例(1.3%),脂膜炎1例(1.3%)。^(18)F-FDG PET/CT诊断为真阳性56例(70.9%),总体诊断敏感度为88.9%、特异度为44.8%、准确度为79.7%、阳性预测值为86.2%、阴性预测值为50.0%。FUO病因类别为感染性疾病(χ^(2)=4.03,P=0.045)或风湿免疫性疾病(χ^(2)=5.64,P=0.018)与^(18)F-FDG PET/CT的诊断贡献显著相关。结论 ^(18)F-FDG PET/CT可作为非侵入性检查,为非肿瘤性疾病导致FUO的临床诊断提供重要参考价值。

某三甲医院门诊不明原因发热患者中感染性因素的列线图预测模型构建与评估

目的 调查某三甲医院综合门诊不明原因发热(FUO)的病因类型,探讨与感染性疾病病因相关的因素,建立预测感染性疾病的列线图模型并评估。方法 回顾性分析2019年1月-2021年6月某三甲医院综合门诊收治的FUO患者445例作为开发队列,收集患者基本特征、实验室检查等,应用LASSO回归优化筛选变量,采用logistic回归模型筛选感染性疾病的独立危险因素并建立预测模型,使用R软件绘制出列线图。内部验证采用Bootstrap法,使用受试者工作特征(ROC)曲线、校准曲线和Hosmer-Lemeshow拟合优度检验验证模型的区分度与校准度,决策曲线分析(DCA)评估模型的临床有效性。另选取2021年7-12月该院综合门诊收治的110例FUO患者作为验证队列对模型进行外部验证。结果 本研究经LASSO回归分析,最终纳入病程<35 d、畏寒/寒战、血清铁蛋白<520μg/L、降钙素原≥0.32μg/L、中性粒细胞<76%及碱性磷酸酶<82 U/L作为FUO患者感染性疾病病因的独立预测因素(P<0.05);依据上述独立预测变量绘制列线图,模型预测开发队列ROC曲线下面积为0.824,最大Youden指数所对应的风险预测值34.62%,预测临界值评分为145分,因此FUO患者得分≥145分的病因很可能是感染性疾病;Hosmer-Lemeshow拟合优度检验χ2=6.223,P=0.514;校准曲线的实际曲线与理想曲线几乎重合;验证队列的ROC曲线下面积为0.768,校准曲线的实际曲线位于理想曲线附近。结论 本研究基于某三甲医院门诊FUO患者的感染性疾病病因的风险因素构建的列线图预测模型具有良好的一致性与区分度,风险预测值评分为145分。