利用聚合酶連鎖反應定量methicillin 抗藥性金黃色葡萄球菌血液中菌量並監測療效

Abstract

過去數十年來，金黃色葡萄球菌(Staphylococcus aureus)在社區及院內感染均有顯著增加，其中methicillin抗藥性金黃色葡萄球菌 (methicillin-resistant S. aureus, MRSA) 為重要原因。MRSA菌血症不僅是一個預後不良的因子，更可能造成持續性的感染，菌血症的復發，住院天數延長及高死亡率；因此，建立有效的新指標來追蹤治療，刻不容緩。隨著分子生物技術的進步，已有研究建立以即時性聚合酶連鎖反應 (real-time polymerase chain reaction, real-time PCR) 用於血液培養瓶及鼻腔拭子的MRSA快速偵測，但目前仍缺乏以此作為MRSA定量及追蹤指標的相關研究。 本研究的第一部份，我們利用Franocois等人所發表的mecA探針及引子建立一即時性聚合酶連鎖反應 (J. Clin. Microbiol. 2003;41:254)，以已知mecA DNA copy number之標準質體 (plasmid standard) 來定量檢體中之未知mecA DNA copy number。我們也建立了femASE的即時性聚合酶連鎖反應來排除常見的汙染菌 – Staphylococcus epidermidis。mecA DNA copy number與已知MRSA菌量呈線性關係 (相關係數 = 0.9967)，與含有已知MRSA菌量之健康成人全血亦呈線性關係 (相關係數 = 0.9995)，每毫升全血之敏感度達100 copies mecA DNA。經四日攝氏4度保存下，mecA DNA copy number維持一定值，證實以攝氏4度保存四日之血液檢體做定量分析仍為可行。 本研究的第二部份，我們利用此即時性聚合酶連鎖反應來定量以血液培養確診為MRSA菌血症的病人，血液中mecA DNA copy number，探討與臨床指標與病程之關係。在這個前瞻性觀察性研究中，在民國九十五年七月一日至九十六年一月三十一日間，於台大醫院加護病房住院病人中，以血液培養確診為MRSA菌血症者，計有20位病人加入本研究，共250個全血檢體，其中87個檢體採檢同時有血液培養同時送檢。其中7位病人死於MRSA菌血症 (死亡組)，13位病人存活 (存活組)。除了死亡組的病人有較高比例曾經接受機械性瓣膜置換術 (43% vs. 0%, P = 0.03)，兩組病人的本身疾病、MRSA感染部位、抗生素之選擇、抗MRSA抗生素給予之時間延誤及感染性導管之移除時間延誤均無顯著差異。 mecA DNA level在血液培養MRSA陽性時較血液培養陰性時高 (1.65x105 vs. 3.49x104 copies/mL, P = 0.00002)，在發燒時較退燒時高 (9.38x104 vs. 2.81x104 copies/mL, P = 0.00005)，在MRSA治療小於三天時較治療大於或等於三天時高(9.26x104 vs. 3.07x104 copies/mL, P = 0.002)。這樣的關係在其他臨床常使用之參考因子，如白血球數及C-reactive protein，則不一定。值得注意的是，血液培養陰性時往往仍能測得mecA DNA level (interquartile range [IQR], 5.45 x103-7.56 x104 copies/mL)。 比較死亡組與存活組，在第一套MRSA陽性血液培養採檢而尚未投予治療時，mecA DNA level兩組間並沒有差異(2.34 x105 vs. 3.98 x104 copies/mL, P = 0.074)。但在MRSA治療後第三天(7.17x105 vs. 2.00 x104 copies/mL, P = 0.02)及第七天(1.62x105 vs 2.31x104 copies/mL, P = 0.04)，死亡組之mecA DNA level較存活組高。死亡組mecA DNA在治療第三天(0.49 vs -0.37 log, P = 0.06)及第七天(0.00 vs -0.42 log, P = 0.07)有上升或持續的趨勢，但未達統計意義。 本研究主要目的在於建立一個即時性聚合酶連鎖反應以定量mecA gene，並以此探討以血液培養確診之MRSA菌血症病人血液中mecA DNA量與臨床指標及病程的關係。本研究發現，此即時性聚合酶反應可用於MRSA菌血症之病程監測，在治療後第三天mecA DNA level可作為評估治療成效之有效新指標。未來以更多病人的不同病程可證實並深入探討本研究之進一步發現。

During the last decade, methicillin-resistant Staphylococcus aureus (MRSA) has become an increasingly important pathogen in both nosocomial and community - acquired infections. The recent report that MRSA bacteremia, a poor prognostic factor for MRSA infections, was responsible for persistent infections, recurrent episodes, prolonged hospital stay and high mortality rate, suggested the importance and the need of developing new methods and good parameters to monitor MRSA bacteremia. With the advancement in molecular technologies, several real-time polymerase chain reaction (PCR) assays were reported to detect MRSA in various clinical samples. However, the possibility of using a quantitative real-time PCR to quantify and monitor MRSA bacteremia has not been explored. For the first specific aim of this study, we have established a quantitative real-time PCR assay for mecA gene by using known copy numbers of a plasmid containing mecA DNA as standard and the previously described mecA specific primers and probes (Francois et al., J. Clin. Microbiol. 2003;41:254). A real-time PCR assay for femASE was also established to exclude the possibility of contamination by Staphylococcus epidermidis. A linear relationship was found between mecA DNA copy numbers detected and the colony forming units (CFU) of a MRSA reference strain (correlation coefficient = 0.9967) and in MRSA-spiked whole blood samples (correlation coefficient = 0.9995). The sensitivity of the assay was estimated to be 100 copies of mecA DNA per mL whole blood. The levels of mecA DNA detected for DNA extracted from MRSA in whole blood stored at 4°C for up to 4 days remain stable, suggesting the feasibility of quantifying mecA DNA derived from MRSA in stored whole blood samples. For the second aim of this study, we employed the quantitative real-time PCR assay for mecA gene to investigate the mecA DNA load in patients with culture-confirmed MRSA bacteremia and its relationships to various clinical parameters and disease outcome. This prospective observational study enrolled patients with culture-proven MRSA bacteremia in the intensive care units of National Taiwan University Hospital between July 1, 2006 and January 31, 2007. A total of 250 blood samples from 20 patients were collected, and 87 blood samples of them had concomitant blood culture performed. Among the 20 patients, 7 patients died of MRSA infection (non-survivor group) and 13 patients survived the episode (survivor group). The underlying illness was not significantly different between the two groups except for a higher percentage of mechanical valve implantation in patients of the non-survivor group than those of the survivor group (43% vs. 0%, P = 0.03). The foci of MRSA infection, the choice and delay of initial anti-MRSA therapy, delay in catheter removal were not statistically different between the two groups. The level of mecA DNA levels were higher in samples with concomitant positive blood culture than those with negative blood cultures (1.65x105 vs. 3.49 x104 copies/mL, P = 0.00002), in samples taken during febrile period than those taken during afebrile period (9.38x104 vs. 2.81x104 copies/mL, P = 0.00005), and samples from patients receiving less than 3 days of anti-MRSA therapy than those from patients receiving >= 3 days of anti-MRSA therapy (9.26x104 vs. 3.07x104 copies/mL, P = 0.002). This trend was variable in other clinical parameters, such as white blood cell counts and C-reactive protein. Of note, the concomitant mecA DNA levels in the blood culture-negative samples were not zero (interquartile range (IQR), 5.45 x103-7.56 x104 copies/mL). On the day of first positive blood culture when the treatment was not initiated, non-survivors and survivors had similar mecA DNA levels (2.34 x105 vs. 3.98 x104 copies/mL, P = 0.07). mecA DNA levels on the third day of treatment (7.17x105 vs. 2.00 x104 copies/mL, P = 0.02) and on the seventh day of treatment (1.62x105 vs 2.31x104 copies/mL, P = 0.04) were higher in the non-survivors than those in the survivors. There was a trend of persistence or increase of mecA DNA levels in the non-survivors and decreasing mecA DNA levels in the survivors at day 3 (0.49 vs -0.37 log, P = 0.06) and day 7 of treatment (0.00 vs -0.42 log, P = 0.07), though this has not reach statistical significance. The overall objective of this study is to develop a quantitative real-time PCR assay for mecA gene and employ this assay to investigate the mecA DNA load in patients with culture-confirmed MRSA bacteremia and its relationships to various clinical parameters and disease outcome. Our study suggested that this mecA real-time PCR assay can be used to monitor MRSA bacteremia at day 3 and day 7 of treatment and has potentials to provide novel and useful parameters to evaluate treatment of MRSA bacteremia. Future investigation with a larger number of patients of MRSA bacteremia with different clinical outcomes would validate and extend the findings from this study.