張歐 林偉浩
[摘要] 目的 探索NNIS評(píng)分系統(tǒng)在預(yù)測(cè)首次及翻修的髖膝關(guān)節(jié)置換術(shù)發(fā)生手術(shù)部位感染的差別。 方法 回顧性分析廣東省某三甲醫(yī)院2014年6月~2019年6月開(kāi)展的1309例髖膝關(guān)節(jié)置換術(shù)病例,采用Logistic回歸比較首次手術(shù)組和翻修手術(shù)組發(fā)生感染的風(fēng)險(xiǎn),并進(jìn)一步探討NNIS評(píng)分分層分析。 結(jié)果 髖膝關(guān)節(jié)置換術(shù)手術(shù)部位感染率為1.83%,其中翻修手術(shù)組感染率為13.43%、首次手術(shù)組感染率為1.21%(OR=12.683);翻修手術(shù)組的手術(shù)時(shí)長(zhǎng)、NNIS評(píng)分、術(shù)后住院和用藥天數(shù)均大于首次手術(shù)組(P<0.05)。以NNIS評(píng)分分層分析,高評(píng)分組發(fā)生手術(shù)部位感染的危險(xiǎn)度是低評(píng)分組的5.3倍;進(jìn)一步以首次和翻修分組,發(fā)現(xiàn)手術(shù)部位感染率NNIS評(píng)分為1的組,差異有統(tǒng)計(jì)學(xué)意義(P=0.000)。多元Logistic回歸分析控制年齡、性別和NNIS評(píng)分后,翻修手術(shù)感染率的危險(xiǎn)度仍然達(dá)到首次手術(shù)的10倍以上。 結(jié)論 未區(qū)分翻修手術(shù)和首次手術(shù)會(huì)導(dǎo)致髖膝關(guān)節(jié)置換術(shù)的手術(shù)部位感染率被高估,NNIS評(píng)分難以預(yù)測(cè)該差異。
[關(guān)鍵詞] NNIS;首次髖膝關(guān)節(jié)置換術(shù);髖膝關(guān)節(jié)翻修術(shù);手術(shù)部位感染
[中圖分類號(hào)] R687.4 ? ? ? ? ?[文獻(xiàn)標(biāo)識(shí)碼] A ? ? ? ? ?[文章編號(hào)] 1673-9701(2020)26-0023-04
[Abstract] Objective To investigate the difference of NNIS scoring system in predicting surgical site infection(SSI) in the first and revision hip and knee joint arthroplasty. Methods A total of 1 309 cases of hip and knee joint arthroplasty performed in a grade 3 and first-class hospital of Guangdong Province from June 2014 to June 2019 were retrospectively analyzed, and logistic regression was used to compare the infection risk between the first arthroplasty group and the revision arthroplasty group, and the NNIS stratified analysis was further explored. Results The SSI rate of hip and knee joint arthroplasty was 1.83%, among which that of the revision arthroplasty group was 13.43% and that of the first arthroplasty group was 1.21%(OR=12.683); the operation duration, the NNIS scores, the postoperative hospitalization and medication days of the revision arthroplasty group were all more than those of the first arthroplasty group(P<0.05). According to stratified analysis of NNIS scores, the risk of SSI in the high score group was 5.3 times that in the low score group; furthermore, according to the first and revision grouping method, the SSI rate was found to be statistically different in the group with NNIS score of 1(P=0.000). Multivariate logistic regression analysis, revealed that after controlling age, sex and NNIS scores, the risk of infection rate of the revision arthroplasty was still 10 times more than that of the first hip and knee joint arthroplasty. Conclusion Failure to distinguish between revision arthroplasty and first arthroplasty will lead to an overestimation of SSI rate of hip and knee joint arthroplasty, which is difficult to predict by NNIS risk index.
[Key words] NNIS; First hip and knee joint arthroplasty; Revision hip and knee joint arthroplasty; Surgical site infection
美國(guó)國(guó)家醫(yī)院感染監(jiān)測(cè)系統(tǒng)(National nosocomial infection surveillance,NNIS)是臨床廣泛應(yīng)用的手術(shù)部位感染(Surgical site infections,SSI)監(jiān)測(cè)工具,該模型通過(guò)對(duì)特定風(fēng)險(xiǎn)指數(shù)分層分析來(lái)預(yù)測(cè)、評(píng)估不同人群的SSI發(fā)生率[1]。但其針對(duì)不同地區(qū)、各類手術(shù)的預(yù)測(cè)適應(yīng)性仍需要深入研究[2-3],如髖膝關(guān)節(jié)翻修術(shù)的SSI危險(xiǎn)度通常大于首次髖膝關(guān)節(jié)置換術(shù),但多數(shù)研究并未深入探討[2]。隨著關(guān)節(jié)置換應(yīng)用增多,術(shù)后需要進(jìn)行翻修手術(shù)的例數(shù)也有所增長(zhǎng),其術(shù)后感染對(duì)患者生存質(zhì)量和醫(yī)療成本造成的不良后果不容忽視[4-5]。本研究收集廣東省某三甲醫(yī)院2014年6月~2019年6月實(shí)施髖膝關(guān)節(jié)置換術(shù)的患者1309例,探討首次髖膝關(guān)節(jié)置換術(shù)和髖膝關(guān)節(jié)翻修術(shù)SSI感染率的差別,并分析NNIS風(fēng)險(xiǎn)指數(shù)對(duì)該術(shù)式的預(yù)測(cè)能力。
1 資料與方法
1.1 一般資料
收集廣東省某三甲醫(yī)院2014年6月~2019年6月髖膝關(guān)節(jié)置換手術(shù)有效病例1309例。
1.2 方法
由醫(yī)院感染管理軟件獲取手術(shù)病例基本人口學(xué)信息,包括姓名、ID、病區(qū)、性別、年齡、出入院日期、術(shù)后3 d最高體溫、術(shù)后抗生素使用天數(shù)和院內(nèi)感染診斷及手術(shù)信息(包括手術(shù)名稱、手術(shù)日期、麻醉方法、愈合等級(jí)、手術(shù)時(shí)間、切口等級(jí)、ASA評(píng)分和NNIS評(píng)分)。
1.3 觀察指標(biāo)
參照2001年原衛(wèi)生部《醫(yī)院感染診斷標(biāo)準(zhǔn)》中手術(shù)部位感染診斷標(biāo)準(zhǔn)。NNIS評(píng)分系統(tǒng)按手術(shù)感染風(fēng)險(xiǎn)分為四級(jí)(分別為0級(jí)、1級(jí)、2級(jí)、3級(jí)),賦值標(biāo)準(zhǔn)為手術(shù)持續(xù)時(shí)間≤75%賦值為0分,>75%賦值為1分;手術(shù)切口分級(jí)≤Ⅱ賦值0分,>Ⅱ賦值1分;ASA分級(jí)≤Ⅱ賦值0分,>Ⅱ賦值1分[2]。由于本研究?jī)H納入Ⅰ類手術(shù)切口,無(wú)病例NNIS評(píng)分為3。
1.4 統(tǒng)計(jì)學(xué)方法
將手術(shù)病例數(shù)據(jù)導(dǎo)入SPSS17.0整理并分析,采用Fisher精確檢驗(yàn)描述性分析髖膝關(guān)節(jié)置換術(shù)病例的人口學(xué)基線資料,分首次手術(shù)和翻修手術(shù)兩組分析統(tǒng)計(jì)學(xué)差異,計(jì)數(shù)資料采用χ2檢驗(yàn)、計(jì)量資料用t檢驗(yàn),P<0.05表示差異有統(tǒng)計(jì)學(xué)意義。采用單變量Logistic回歸分析NNIS評(píng)分與SSI感染率的關(guān)系,并將相關(guān)自變量納入多元Logistic回歸比較兩組發(fā)生SSI的危險(xiǎn)度差異,因ASA分級(jí)和手術(shù)時(shí)長(zhǎng)與NNIS評(píng)分存在共線性被排除在多元模型外。
2 結(jié)果
2.1 髖膝關(guān)節(jié)置換術(shù)患者一般資料
本次研究納入近5年髖膝關(guān)節(jié)置換術(shù)病例1309例,均為Ⅰ類手術(shù)切口(清潔切口),其中髖膝關(guān)節(jié)翻修手術(shù)67例。首次手術(shù)和翻修手術(shù)兩組患者的年齡和性別分布均衡(P>0.05),體現(xiàn)患者體質(zhì)狀況的ASA評(píng)分也沒(méi)有統(tǒng)計(jì)學(xué)差異(P>0.05)。兩組手術(shù)時(shí)長(zhǎng)和NNIS評(píng)分差異顯著(P<0.05),翻修手術(shù)組均大于首次手術(shù)組。翻修手術(shù)組術(shù)后住院天數(shù)和用藥天數(shù)均顯著高于首次手術(shù)組(P<0.05)。見(jiàn)表1。
1309例髖膝關(guān)節(jié)置換術(shù)中有24例發(fā)生手術(shù)部位感染,SSI感染率1.83%,首次手術(shù)組感染率1.21%、翻修手術(shù)組感染率13.43%[OR=12.683(95%置信區(qū)間,5.328~30.192)]。按感染部位分,9例(37.5%)發(fā)生表淺手術(shù)切口感染、5例(20.8%)深部手術(shù)切口感染、10例(41.7%)器官腔隙感染。按NNIS評(píng)分分層,SSI感染率依次為0.98%(0分)、1.35%(1分)、15.52%(2分),單變量Logistic回歸分析顯示NNIS評(píng)分高的組發(fā)生SSI的危險(xiǎn)度更高[OR=5.169(95%置信區(qū)間,2.531~11.024)]。見(jiàn)表2。
2.3 首次手術(shù)組與翻修手術(shù)組SSI感染率的關(guān)系
1309例髖膝關(guān)節(jié)置換術(shù)中有67例翻修手術(shù),9例感染(13.4%);其中深部切口和器官腔隙感染例數(shù)為7例(77.8%)。經(jīng)NNIS評(píng)分分層后,首次手術(shù)和翻修手術(shù)手術(shù)部位感染率的差異在NNIS評(píng)分為1的組出現(xiàn)了統(tǒng)計(jì)學(xué)差異(P<0.05,圖1);當(dāng)感染病例限制在深部切口和器官腔隙感染時(shí),NNIS=1組的SSI感染率也有統(tǒng)計(jì)學(xué)差異(P<0.05)。多元logistic回歸分析在控制了年齡、性別和NNIS評(píng)分后,翻修手術(shù)手術(shù)部位感染的危險(xiǎn)性仍然達(dá)到首次手術(shù)的10倍以上[OR=10.2(95%置信區(qū)間3.9~26.1)]。
3討論
我國(guó)從2010年開(kāi)始廣泛應(yīng)用NNIS評(píng)分系統(tǒng)預(yù)測(cè)手術(shù)部位感染(SSI),該系統(tǒng)是美國(guó)在1991年通過(guò)Logistic回歸模型篩選,確定了美國(guó)麻醉醫(yī)師學(xué)會(huì)(ASA)身體狀態(tài)分級(jí)、手術(shù)切口類型、手術(shù)持續(xù)時(shí)間(T)三個(gè)風(fēng)險(xiǎn)指數(shù)[1]。但有研究顯示其在美國(guó)以外某些地區(qū)預(yù)測(cè)敏感性較低[3,5],且對(duì)不同術(shù)式SSI感染率的預(yù)測(cè)能力有差異[6]。隨著老齡化人口和高能量損傷年輕患者的不斷增加,關(guān)節(jié)置換的應(yīng)用越來(lái)越多,其手術(shù)感染并發(fā)癥的監(jiān)測(cè)評(píng)估也備受關(guān)注[4]。本次研究旨在探索NNIS評(píng)分系統(tǒng)在首次髖膝關(guān)節(jié)置換術(shù)和髖膝關(guān)節(jié)翻修術(shù)發(fā)生SSI的預(yù)測(cè)能力,并評(píng)估這兩類手術(shù)感染率的差異。
由于翻修手術(shù)組納入的病例數(shù)有限,本研究的基線資料分析采用了Fisher精確檢驗(yàn),結(jié)果顯示首次手術(shù)和翻修手術(shù)的兩組患者基本情況分布均衡。但預(yù)后有顯著性差異,兩組的手術(shù)時(shí)長(zhǎng)、NNIS評(píng)分、術(shù)后住院天數(shù)和用藥天數(shù)均有統(tǒng)計(jì)學(xué)差異,且經(jīng)單變量分析發(fā)現(xiàn)翻修手術(shù)組發(fā)生SSI的危險(xiǎn)度是首次手術(shù)組的10倍以上。這體現(xiàn)了首次手術(shù)組和翻修手術(shù)組的手術(shù)復(fù)雜度、技術(shù)要求、環(huán)境微生物暴露時(shí)長(zhǎng)、抗生素的效力、預(yù)后和醫(yī)療成本的差別[7]。
本研究以NNIS評(píng)分進(jìn)行分層分析,以探索NNIS評(píng)分系統(tǒng)對(duì)不同危險(xiǎn)度手術(shù)的預(yù)測(cè)能力及穩(wěn)定性。結(jié)果顯示,在NNIS評(píng)分為1分時(shí)首次手術(shù)和翻修手術(shù)兩組的SSI感染率有統(tǒng)計(jì)學(xué)差異,說(shuō)明兩組病例人群非同質(zhì),NNIS評(píng)分為0、2分時(shí)無(wú)顯著性差異,考慮可能因?yàn)閮山M數(shù)據(jù)相對(duì)較少。何文英[1]的研究通過(guò)ROC曲線下面積(AUC)評(píng)價(jià)NNIS評(píng)分對(duì)髖膝關(guān)節(jié)置換術(shù)SSI的預(yù)測(cè)能力為0.608(AUC=0.5,表明無(wú)預(yù)測(cè)價(jià)值),考慮是因?yàn)樵擃愂中g(shù)的感染率低,稀釋了NNIS評(píng)分的預(yù)測(cè)能力。NNIS評(píng)分系統(tǒng)預(yù)測(cè)能力的局限性也相似地體現(xiàn)在其他感染率低的術(shù)式,如糖尿病和肥胖對(duì)冠狀動(dòng)脈搭橋術(shù)的影響就大于NNIS評(píng)分[8-10]。有研究顯示NNIS評(píng)分更適合易感染的術(shù)式,且推薦出現(xiàn)復(fù)雜感染的病例(如深部切口和器官腔隙感染)使用NNIS評(píng)分[11-12]。于是本研究進(jìn)一步探索發(fā)生深部切口感染和器官腔隙感染這類復(fù)雜感染的病例,發(fā)現(xiàn)評(píng)分為1組的感染率仍有統(tǒng)計(jì)學(xué)差異。
在此基礎(chǔ)上,本研究采用多元Logistic回歸模型進(jìn)一步探討,NNIS評(píng)分作為監(jiān)測(cè)工具是否能體現(xiàn)出首次手術(shù)和翻修手術(shù)患者術(shù)后出現(xiàn)SSI感染的差別。通過(guò)多元回歸控制了年齡、性別、NNIS評(píng)分后,翻修手術(shù)組的SSI危險(xiǎn)度仍高于首次手術(shù)組,且達(dá)到10倍以上,說(shuō)明NNIS評(píng)分系數(shù)體現(xiàn)不出首次手術(shù)與翻修手術(shù)發(fā)生SSI的區(qū)別,即兩組患者不同質(zhì)??赡芤?yàn)榛颊呓?jīng)歷初次置換術(shù)后,如果出現(xiàn)手術(shù)部位感染、假體松動(dòng)、假體周圍骨折、聚乙烯磨損所致骨缺損等情況,會(huì)進(jìn)行二次翻修,故翻修手術(shù)一般意味著較長(zhǎng)的手術(shù)時(shí)間、更復(fù)雜的術(shù)式和更差的患者狀態(tài)[13-15],進(jìn)而導(dǎo)致兩組患者的本質(zhì)差別。本研究已排除因首次手術(shù)后發(fā)生SSI而進(jìn)行翻修手術(shù)的病例。
本研究回顧性分析了近5年的髖膝關(guān)節(jié)置換術(shù)病例,數(shù)據(jù)量較大;并從NNIS評(píng)分系統(tǒng)監(jiān)測(cè)能力的角度探索了首次手術(shù)和翻修手術(shù)發(fā)生SSI的差別;也明確了監(jiān)測(cè)髖膝關(guān)節(jié)置換術(shù)SSI感染率或進(jìn)行危險(xiǎn)因素分析時(shí),應(yīng)區(qū)分首次手術(shù)和翻修手術(shù),否則會(huì)造成對(duì)手術(shù)部位感染率的高估。不足之處在于,欠缺全病例隨訪可能會(huì)造成部分SSI陽(yáng)性病例遺失,但該問(wèn)題在翻修手術(shù)組和首次手術(shù)組是均衡的,不會(huì)對(duì)現(xiàn)有結(jié)果造成太大影響??傊狙芯刻崾綨NIS評(píng)分應(yīng)用的局限性,為今后進(jìn)行手術(shù)部位感染的預(yù)測(cè)研究提供了不同的思路。
[參考文獻(xiàn)]
[1] 何文英,張炎,黃新玲,等.NNIS風(fēng)險(xiǎn)指數(shù)在手術(shù)部位感染應(yīng)用的效果評(píng)估[J].中華醫(yī)院感染學(xué)雜志,2017, 27(8):1832-1836.
[2] 杜輝,尹星華,黃勇,等.人工髖和膝關(guān)節(jié)置換術(shù)后患者住院期間病死率及病死的危險(xiǎn)因素分析[J].骨科臨床與研究雜志,2018,3(5):289-294.
[3] Brennan-Olsen SL,Vogrin S,Graves S,et al.Revision joint replacement surgeries of the hip and knee across geographic region and socioeconomic status in the western region of Victoria:A cross-sectional multilevel analysis of registry data[J].BMC Musculoskeletal Disorders,2019,20(1):300-306.
[4] Tay K,Tang A,F(xiàn)ary C,et al.The effect of surgical approach on early complications of total hip arthoplasty[J].Arthroplasty,2019,1(1):1-7.
[5] 彭慧明,翁習(xí)生.人工髖、膝關(guān)節(jié)置換術(shù)應(yīng)用的原則與策略[J].骨科臨床與研究雜志,2019,4(6):321-324.
[6] Millstone DB,Perruccio AV,Badley EM,et al.Factors associated with adverse events in inpatient elective spine,knee,and hip orthopaedic surgery[J].The Journal of Bone and Joint Surgery,2017,99(16):1365-1372.
[7] Samaniego AR,Gaviria PE,Pons CM,et al.Arthroplasty knee registry of Catalonia:What scientific evidence supports the implantation of our prosthesis?[J].Revista Espaola de Cirugía Ortopédicay Traumatología(English Edition),2018,62(4):290-296.
[8] Lenguerrand E,Beswick AD,Whitehouse MR,et al.Outcomes following hip and knee replacement in diabetic versus nondiabetic patients and well versus poorly controlled diabetic patients:A prospective cohort study[J].Acta Orthopaedica,2018,89(12):1-7.
[9] 劉玉寶,史冬泉,徐興全,等.初次髖膝關(guān)節(jié)置換術(shù)后假體周圍感染危險(xiǎn)因素分析[J].中華關(guān)節(jié)外科雜志(電子版),2017,11(6):37-43.
[10] Agnes K,Mcgrail L,Johnson,et al.Postoperative deep surgical site infections in orthopedic surgeries[J].Ajic American Journal of Infection Control,2004,32(3):107-108.
[11] 孫彬.人工關(guān)節(jié)置換術(shù)后假體周圍感染診斷的研究進(jìn)展[J].中國(guó)微創(chuàng)外科雜志,2019,19(5):439-444.
[12] Eneqvist T,Szilárd N,Erik B,et al.Can patient-reported outcomes predict re-operations after total hip replacement?[J].International Orthopaedics,2018,42(2):1-7.
[13] Sanchez-Santos MT,Garriga C,Judge A,et al.Development and validation of a clinical prediction model of patient-reported pain and function after primary total knee replacement surgery[J].Scientific Reports,2018,8(1):3381.
[14] Robinson PD,Mcewan J,Adukia V,et al.Osteoarthritis and arthroplasty of the hip and knee[J].British Journal of Hospital Medicine,2018,79(4):C54-C59.
[15] Cordtz RL,Zobbe K,Pil H,et al.Predictors of revision,prosthetic joint infection and mortality following total hip or total knee arthroplasty in patients with rheumatoid arthritis:A nationwide cohort study using Danish healthcare registers[J].Annals of the Rheumatic Diseases,2018, 77(2):281-286.
(收稿日期:2020-05-12)