※ 본지는 의학계 발전을 위해 고군분투하는 의학자들의 연구의지를 고취하고 우수한 연구성과를 공유해 보다 많은 의료진과 의학계 관계자들에게 질환으로 고통받고 있는 환자의 진단과 치료는 물론 예방에 이르기까지 긍정적이고 바람직한 영향을 미칠수 있도록 매월 최근 연구논문 중 이 같은 목적에 부합하는 의미를 갖는 논문을 선정해 커버스토리로 논문전문과 저자의 연구 배경. 연구 과정과 성과를 게재한다.
[헬스앤라이프 김세영 기자] 헬스앤라이프저널이 선정한 2019년 5월호 <이달의 논문>은 기승정 전남대병원 진단검사의학과 교수가 지난 3월 15일 세계적인 호흡기 학술지 <미국호흡기 및 중환자의학(American Journal of Respiratory and Critical Care Medicine: IF 15.24)>에 게재한 'Xpert MTB/RIF Assay as a Substitute for Smear Microscopy in an Intermediate-Burden Setting'이다.
‘결핵 중위험 지역에서 현미경 항산균 도말검사의 대안으로써 Xpert 결핵/리팜핀 내성검사(Xpert MTB/RIF Assay as a Substitute for Smear Microscopy in an Intermediate-Burden Setting: 제1저자 이현승 전문의, 교신저자 기승정 교수)’란 제목으로 게재됐다. 기승정 교수는 지금까지 객담을 슬라이드에 얇게 펴 현미경으로 결핵균을 관찰하는 도말검사보다 최신 Xpert 검사법이 훨씬 빠르고 정확하다는 사실을 이번 연구를 통해 입증했다. 연구는 지난 2014년부터 2016년까지 전남대병원 호흡기내과의 폐결핵 의심환자 3000여 명을 대상으로 했다.
Xpert MTB/RIF Assay as a Substitute for Smear Microscopy in an Intermediate-Burden Setting
Hyun-Seung Lee1, Seung-Jung Kee2*, Ju-Hyeon Shin2, Yong-Soo Kwon3, Sejong Chun2, Jun Hyung Lee4,
Eun Jeong Won2,5, Hyun-Jung Choi4, Soo Hyun Kim4, Myung-Geun Shin4, Jong-Hee Shin2, and Soon-Pal Suh2*
At a Glance Commentary
Scientific Knowledge on the Subject: Data about the performance of Xpert MTB/RIF and smear microscopy according to sample-related factors, quantitative capability, and compliance to the turnaround time within 24 hours in routine clinical practice are limited in an intermediate tuberculosis-burden setting.
What This Study Adds to the Field: Sampling time and location significantly affect the performance of smear microscopy, but not the superior performance of Xpert MTB/RIF. An interesting new finding is that the correlation between Xpert cycle threshold value and smear grade was higher than that between Xpert cycle threshold value and time to culture positivity. However, these correlations were similar to that between smear grade and time to culture positivity. Xpert MTB/RIF-based strategy shortens the time to report for diagnosis of pulmonary tuberculosis: in almost all samples, the time to report for Xpert MTB/RIF was shorter than that for smear microscopy.
Furthermore, the compliance of Xpert MTB/RIF to less than 24-hour-long time to report was better than that of smear microscopy. These data indicate that the Xpert MTB/RIF is a more rapid and superior substitute for smear microscopy for use as a primary tool for diagnosis of pulmonary tuberculosis, irrespective of sample-related factors, and complementary to smear microscopy as a measure of bacterial burden in routine clinical practice in intermediate-burden areas.
Tuberculosis, caused by Mycobacterium tuberculosis complex, is a major global public concern with 10.4 million new cases and 1.3 million deaths worldwide in 2017 (1). Rapid and accurate diagnosis is critical for timely control of tuberculosis. The most commonly rapid tuberculosis test is sputum smear microscopy, which dates back to 1882, when Robert Koch first reported that he developed the staining technique that allowed him to see the tubercle bacillus under a microscope. The staining technique that was quickly refined by Paul Erlich and then further modified by Ziehl and Neelsen is basically still used today (2). Xpert MTB/RIF assay (Xpert: Cepheid, Sunnyvale,CA), another rapid tuberculosis test, which has beenpolicy-endorsed by World Health Organization in 2011, isa fully automated cartridge-based molecular method forsimultaneous detection of Mycobacterium tuberculosis complex and rifampin resistance (3). In an editorial describing the outcome of implementation research of a streamlined strategy for improving tuberculosis diagnosis and treatment for patients receiving care at five district government primary care health centers in Uganda, Robert Belknap and Rendall Reves wrote, “It is time to garner the resources to move beyond the smear microscopy as the initial diagnostic tests and toward the goal of treatment based upon universal drug susceptibility testing. We are hopeful that future studies will evaluate Xpert and/or other nucleic acid amplification tests as replacements for acid-fast bacilli smears” (4).
Shortly thereafter, in the recent editorial describing our study, Neil W. Schluger wrote, “The article by Lee and colleagues provides further evidence that it is time for smear microscopy to find its place in medical museums
and history books, rather than modern diagnostic labs” (2).
Xpert has high sensitivity, requiring 131 colony-forming units per milliliter, whereas smear microscopy has low sensitivity, requiring the presence of 5,000-10,000 bacilli per milliliter of sample to allow detection (5,6). Furthermore, compared with Xpert assay, smear microscopy has lower specificity, especially in patients with nontuberculous mycobacteria lung disease who are an increasing proportion of patients with smear-positive lung disease in lower tuberculosis-burden populations (7). Furthermore, Xpert provides semiquantitative results based on the cycle-threshold (CT) of the first positive probe that detects Mycobacterium tuberculosis complex, which is correlated with the mycobacterial load in sputum samples as defined by smear grade and time to culture positivity (8, 9). According to the Centers for Disease Control and Prevention guidelines, the results of smear microscopy should be reported within 24 hours of sample collection (6). Xpert usually provides results within 2 hours of sample collection, thereby reducing the time period before initiating anti-tuberculosis therapy (10, 11). However, use of Xpert assay as a substitute for smear microscopy in routine clinical practice remains unexplored in an intermediate tuberculosis-burden setting.
Accordingly, the aim of this study was to compare the diagnostic performance of Xpert and smear microscopy, based on sampling time and location, correlation of Xpert semiquantitative category with smear grade and time-to-culture positivity, and compliance of reporting time with defined standard time for routine clinical use in an intermediate TB-burden setting.
A total of 3,359 consecutive sputum samples were collected from 2,952 adult patients with suspected pulmonary TB. One sputum sample was directly tested by Xpert. Briefly, 1mL of sputum sample and 2 mL of the reagent were mixed and incubated for about 10 min at room temperature. The incubated samples were lodged into the artridges, and the tests were performed automatically. Another sputum sample was pretreated with equal volumes of 4% sodium hydroxide and centrifuged at 3000 g for 20 min. One aliquot of sediment was screened by rhodamine auramine fluorescence staining and confirmed by Ziehl–Neelsen staining. Two were used for culture on 2% Ogawa medium(Asan Pharmaceutical Co., Seoul, Korea) as well as in BACTEC MGIT liquid medium (Becton Dickinson Diagnostic Instrument Systems, Sparks, MD).
To the best of our knowledge, this is the most extensive single-center study to investigate the performance, quantitative capability, and time to report for Xpert for routine diagnosis of tuberculosis in a large set of consecutive patients with suspected pulmonary tuberculosis in an intermediate tuberculosis-burden setting. The clinical and demographic characteristics of the patients with suspected pulmonary tuberculosis are summarized in Table 1. Of 2,952 patients, 263 (8.9%) were culture-positive for Mycobacterium tuberculosis complex, including one patient carrying a mixed culture of Mycobacterium tuberculosis complex and nontuberculous mycobacteria, of which were 102 smear-positive and 161 smear-negative; 265 (9.0%) were culture-positive for only nontuberculous mycobacteria, consisting of 82 smear-positive and 183 smear-negative; and 2,424 (82.1%) did not show growth in mycobacterial culture, consisting of 8 smear-positive and 2,416 smear-negative. Thus, among a total of 184 smear-positive specimens that grew mycobacteria, 82 (44.6%) were proven nontuberculous mycobacteria (Figure 1).
The diagnostic performance of Xpert and smear microscopy according to heterogeneity factors is summarized in Table2. Using mycobacterial culture as the reference standard, the overall sensitivity of Xpert was significantly higher compared with smear microscopy (74.1% versus 38.8%, P<0.0001). Specificity was comparable between Xpert and smear microscopy (97.5% versus 96.7%, P>0.05). In the 265 nontuberculous mycobacteria-positive samples, Xpert was found to be more specific than smear microscopy (98.9% versus 69.1%, P<0.0001). Additional analysis of heterogeneity factors, including sampling time, sample number, or sampling location, demonstrated that Xpert performed better than smear microscopy. Furthermore, smear microscopy displayed higher diagnostic performance only in early morning and inpatient samples compared to spot and outpatient samples, respectively. However, the diagnostic performance of Xpert was not affected by these heterogeneity factors. Taken together, our findings suggest that the performance of Xpert is more stable and superior to smear microscopy, irrespective of heterogeneity.
The correlation among Xpert semiquantitative category, smear grade, and time to culture positivity is presented in Figure 2. The analysis showed that Xpert category significantly correlated with smear grade (γGoodman-Kruskal= 0.982, P<0.0001); Xpert category significantly correlated with TTP in weeks (γGoodman-Kruskal= -0.962, P<0.0001); and smear grade significantly correlated with TTP in weeks (γGoodman-Kruskal= -0.982, P<0.0001). Time to report for Xpert and smear microscopy is summarized in Table 3. The median time to report for Xpert was remarkably shorter than that of smear microscopy, other real-time polymerase chain reaction assays, and MGIT culture. Notably, Bland-Altman analysis showed that Xpert assay offers faster reporting, compared to smear microscopy, in almost all individual cases. In our study, Xpert displayed a better time to report compliance rate within 24 hours than smear microscopy.
In conclusion, our study showed that the accuracy of Xpert assay for diagnosing pulmonary tuberculosis was profoundly superior to smear microscopy, irrespective of the heterogeneity in clinical subgroups. Furthermore, our study showed that the Xpert semiquantitative level strongly correlated with the smear grade or time to culture positivity with respect to the prediction of mycobacterial burden. Impressively, in almost all cases, the assay offered more rapid reporting, compared to smear microscopy. Taken together, these findings suggest that Xpert can replace smear microscopy as the primary tool for diagnosis of pulmonary tuberculosis patients in routine clinical practice in intermediate tuberculosis-burden areas.
※ 논문 요약 및 표, 그림 자료 : 저자 기승정 전남대병원 교수
#헬스앤라이프 #저널 #이달의논문 #결핵검사 #검사법비교 #초기검사 #기승정 #전남대병원 #진단검사의학과 #Xpert #XpertMTB/RIF #smear microscopy #김세영기자