呼吸衰竭，持续发热和肺泡灌洗液CMV阳性：真实的结果，还是并无关系？

Rita Pechulis MD & Samuel Krachman DO

Temple University

上期问题答案

问题1：此患者支气管灌洗液中针对CMV的最佳检测方法是？

1. CMV培养

2. 快速离心分离病毒(rapid shell vial assay)

3. 抗原血症检测

4. 血清学检查(IgG, IgM滴度)

5. PCR

答案：2. 快速离心分离病毒

这例患者CMV快速离心分离病毒结果为阳性。这种检验方法可用于多种临床标本，如尿液、血液和支气管灌洗液。进行检测时，需要对标本进行离心，以增加病毒吸附；然后单层细胞暴露于MIEp72抗原的特异性单克隆荧光抗体。如果有抗体结合即表明在细胞内有CMV早期复制，一般在24小时内就可以获得结果。

快速离心分离病毒的主要局限性在于敏感性较低。采用CMV培养作为金标准，快速离心分离病毒的敏感性69%，特异性96%[1]。然而，重要的是了解标本中存在CMV不一定提示存在活动性感染；诊断CMV感染必须结合临床表现、体格检查和病毒检测进行综合考虑。

CMV培养很少用于诊断活动性感染。CMV生长缓慢，需要1-6周才可引起明显的细胞病变。而且，即使存在CMV病毒血症，也不能确定存在活动性感染，这是因为在原发感染后，无症状的病毒脱落(shedding)仍可持续数月。由于存在上述局限性，因而很少进行CMV培养。但是，CMV培养可用于确定病毒耐药情况。

PP65是CMV的一种结构性蛋白，抗原血症检测能够定量检测PP65阳性的白细胞数目。阳性结果报告的方式为经过计数的细胞总数中染色阳性的细胞数。一项有关合并CMV感染的HIV阳性患者的研究表明，对于原发感染，抗原血症检测的敏感性97%，对复发感染的敏感性77%，总的敏感性91%，特异性73%[2]。阴性结果有助于排除原发性CMV感染，但在确定是否为活动性疾病时特异性仍显不足。

通过血清学检查诊断CMV感染包括检测CMV特异性IgM抗体，以及观察到相隔2-4周的标本中CMV特异性IgG滴度呈至少4倍升高。这种检测方法的局限性包括需要相隔2-4周采取血清标本。而且，IgM抗体可持续数月，因此造成急性感染诊断的困难。然而，抗体滴度能够确诊既往的CMV感染或暴露，可能有助于IgG抗体滴度阳性患者的预防。

还有多种分子检测方法，如定量和定性PCR、杂交捕获和核酸序列依赖性扩增法。定量和定性PCR能够检测全血、白细胞和血浆中的病毒DNA。对于AIDS患者，定性PCR诊断CMV感染的敏感性和特异性分别为89%和75%。定量PCR可用于监测病情变化，因为AIDS患者的病毒DNA高载量与临床症状相关。杂交捕获是应用一种针对CMV基因组17%核酸的RNA探针，以RNA：DNA杂交结合的方式使样本与抗体结合。这种检验方法可采用全血标本进行检测。核酸序列依赖性扩增法可检测到CMV感染的即刻早期基因UL123（IE1）和晚期基因（pp67）表达，可以在DNA的背景下对非结合的病毒mRNA进行扩增。这种检验方法也可用于全血标本的检测[3]。

本期问题

Adjunctive therapy consists of several approaches to inhibit pathomechanisms of severe sepsis.

This is a new recommendation since 2008 because a number of clinical studies have verified that formal screening protocols for sepsis improve outcome (Jones et al.JAMA. 2010;303:739-746; Moore et al. J Trauma. 2009;66:1539-1546).

The underlined sections are new to the 2012 guidelines as ample evidence now exists to recommend these, rather than simple suggestions. (Memel L, Maki D. Ann Intern Med.1993;119:270-272).

The underlined recommendation on antigen detection systems for candidemia is new. It is based upon a number of clinical studies in neutropenic patients and a meta-analysis that included some ICU patients showing these assays, despite suboptimal sensitivity and specificity, allow for significantly faster diagnosis of invasive candidiasis and earlier intervention with antifungal agents. Early therapy for candida sepsis improves outcome (Alam et al. BMC Infect Dis. 2007;7:103; Sendid et al. Clin Vaccine Immunol. 2008;15:1868-1877).

This recommendation was not graded in the 2012 guidelines as it is based on clinical experience and expert opinion rather than specific clinical trials.

The 1-hour recommendation for severe sepsis without septic shock was ungraded in the 2008 guidelines as some additional studies have suggested early intervention is an important prognostic indicator even without shock. There are fewer data here than in patients with septic shock. The next slide provides a summary of the evidence. The remark was added to clarify that the guidelines subcommittee acknowledges that this is a goal that may be difficult, if not impossible, to fully implement in all healthcare settings, busy emergency departments, and mass casualty settings, but should be the goal where possible.

A summary of recent studies supporting broad-spectrum, empiric combination therapy in the early treatment of sepsis/septic shock.

Legend s-sepsis; ss-severe sepsis; SS-septic shock; HD-hospital discharge; ED-emergency department; or-odds ratio; ns-not significant; HR-hazard ratio; MVR-multivariate regression

1.Kumar et al. Crit Care Med. 2006;34:1589–1596

2.Barochia et al. Crit Care Med. 2010;38:668–678

3.Ferrer et al. Am J Respir Crit Care Med. 2009;180:861–866

4.Barie et al. Surg Infect. 2005;6:41–54

5.Levy et al. Crit Care Med. 2010;38:367–374

6.El Sohl et al. J Am Geriat Soc. 2008;56:272–278

7.Gurnani et al. Clin Ther. 2010;32:1285–1293

8.Nguyen et al. Crit Care Med. 2007;35:1105–1112

9.Castellanos-Ortega et al. Crit Care Med.2010;38:1036–1043

10.Gaieski et al. Crit Care Med. 2010;38:801-807

11.Larsen et al. Pediatrics. 2011;127: e1585–e1592

These underlined changes from the 2008 guidelines were added to acknowledge that viral pathogens (e.g., influenza, hemorrhagic fever viruses) can cause sepsis on occasion and that antibiotic choices need to consider the levels of antibiotics achievable at the site of infection (cerebrospinal fluid, intracellular spaces, when appropriate) [Smith et al. Crit Care Med.2010;38:41-51]. The de-escalation comment and the recommendation were upgraded to Grade 1B from the 2008 guidelines (Grade 1C) as evidence of reduced risk of antibiotic resistance can be accomplished by careful use and antibiotic stewardship to retain the activity of available antibiotics.

• Gao F et al. Crit Care.2005;9:R764–R770

• Schorr C. Crit Care Clin.2009;25:857–867

• Girardis M et al. Crit Care.2009;13:R143

• Pestaña D et al. J Trauma.2010;69:1282–1287

• Berenholtz SM et al. Jt Comm J Qual Patient Safety.2007;33:559–568

• Masterton RG. Crit Care Clin. 2011;27:149-162

This is a new recommendation in the 2012 guidelines. Procalcitonin (or other biomarkers) has been studied in a number of trials as a guide to discontinue potentially unnecessary empirical antibiotics when the clinical situation is stabilized. The available data suggest that this might be useful in saving money and reducing antibiotic use (Heyland DK et al. Crit Care Med.2011;39:1792-1799). However, the safety of this measure and the likelihood it can significantly reduce antibiotic resistance development of Clostridium difficile or other adverse events associated with continued antibiotic use remains to be demonstrated. Procalcitonin levels are not recommended as a biomarker for the diagnosis of sepsis.

We upgraded the recommendation for neutropenia and for Pseudomonas spp. severe sepsis to 2B from 2D in the 2008 guidelines and added other difficult to treat pathogens such as Acinetobacter spp. in the 2012 guidelines. It has been repeatedly demonstrated that empiric therapy that does not cover these pathogens that are later found to be the cause of the infection is associated with significantly worse outcomes. However, much of this clinical data is not based on ICU patients with septic shock, and this indirectness of the data leads us to make it a suggestion (Grade 2) rather than a Grade 1 recommendation. The next two slides provide further supportive evidence of the value of empiric combination antibiotics in treating septic shock to assure that the infecting microorganism that caused septic shock is treated (Kumar et al. Crit Care Med. 2010;38:1773-1785; Micek ST et al. Antimicrob Agents Chemother.2010;54:1742-1748).

Kumar et al. Crit Care Med. 2010;38:1773–1785.

Legend: HR-hazard ratio; CI-confidence interval; or-odds ratio.

Micek et al. Antimicrob Agents Chemother. 2010;54:1742–1748.

Legend: IIAT-inappropriate initial antimicrobial therapy, AOR-adjusted odds ratio, CI-confidence interval, APACHE-Acute Physiology and Chronic Health Evaluation.

Kumar et al. Crit Care Med. 2010;38:1651–1664

Garnacho-Montero et al. Crit Care Med. 2007;35:1888-1895

Legend: or-odds ratio; CI-confidence interval; VAP-ventilator-associated pneumonia; AHR-adjusted hazards ratio.

Al-Hasan et al. Antimicrob Agents Chemother. 2009;53:1386-1394.

Legend: GNB=Gram-negative bacteria, AHR-adjusted hazard ratio, CI-confidence interval.

Martinez et al. Antimicrob Agents Chemother. 2010;54:3590-3596

Rodriguez et al. Crit Care Med. 2007; 35:1493–1498

Legend: GNB-Gram-negative bacteria, or-odds ratio, CI-confidence interval, ESBL-extended-spectrum beta-lactamase, CAP-community-acquired pneumonia, HR-hazard ratio.

This recommendation relating to pneumococcal septic shock was added to the 2012 guidelines in keeping with a number of studies (see next slide for summary of evidence) suggesting improved outcomes for patients with septic shock when a macrolide is added to a beta-lactam in the early phase of treatment. The reasons for this apparent improved outcome in some, but not all, clinical studies are not clear but might relate to the anti-inflammatory actions attributable to macrolides or other mechanisms. It was decided to make this a suggestion until further clinical studies can support or refute this evidence.

The collective, observational clinical data support adding a macrolide to beta-lactams even in documented, bacteremic pneumococcal septic shock. The mechanism responsible for this apparent benefit is unclear but is generally attributable to the anti-inflammatory effects of macrolides. Combination therapy is a level 2 recommendation, despite the additional expense and concern over possible promotion of macrolide antibiotic resistance by its widespread, potentially unnecessary use for this indication. Further study and mechanistic evidence to support its use is clearly needed to support combination treatment over standard beta-lactam monotherapy.

Legend: PSI-pneumonia severity index, B-bacteremia, S-sepsis, MV-mechanical ventilation, ns-not significant.

• Martinez et al. Clin Infect Dis.2003;36:389-395

• Baddour et al. Am J Respir Crit Care Med.2004;170:440-444

• Weiss et al. Can Respir J.2004;11:589-593

• Rodriguez et al. Crit Care Med.2007;35:1493–1498

• Lodise et al. Antimicrob Agents Chemother.2007;51:3977-3982

• Chokshi et al. Eur J Clin Microbiol Infect Dis.2007;26:447-451

• Martin-Loesches et al. Intensive Care Med.2010;36:612–620

The grade was increased from 2D in 2008 to 2B in 2012 as some new clinical data on early de-escalation of antibiotics suggest that this can be done safely. Exceptions exist to rapid conversion to monotherapy as indicated. Useful references include:

• Schorr C. Crit Care Clin.2009;25:857–867

• Girardis M. Crit Care.2009;13:R143

• Pestaña D. J Trauma.2010;69:1282–1287

• Berenholtz SM. Jt Comm J Qual Patient Saf.2007;33:559–568

• Black MD. Crit Care Med.2012;40:1324–1328

• Masterton RG. Crit CareClin.201;27:149-162

The 2008 guidelines did not contain this recommendation. The committee wanted to clarify that certain specific indications for antibiotic therapy require more prolonged therapy than 7 to 10 days to avoid antibiotic failures and relapsing infections. Some of those situations are listed here and are provided as suggestions rather recommendations (Grade 2C).

These are new guideline statements following the events of the 2009 influenza pandemic when evidence rapidly accrued that early institution with antiviral therapy could be lifesaving in severe influenza. The second point was added to point out that antibiotics should be stopped as soon as possible after it is determined that an infectious disease is not causing the acute inflammatory state that was suspected to be due to sepsis. Preserving antibiotics until their use is warranted is important in ICU care and in medical care in general as part of antibiotic stewardship programs.

These recommendations and grade guidelines are comparable to those of 2008 except the time interval was lengthened for practical reasons to 12 hours versus the original 6-hour period.

These recommendations were not graded as they were in the 2008 guidelines as they are primarily based upon common sense clinical observations and expert opinion rather than high-level clinical study evidence.

The 2008 guidelines did not make a selective digestive decontamination recommendation. Despite the ongoing controversy over its efficacy and impact on antibiotic resistance patterns, the committee agreed that it should be graded as much evidence now exists in the medical literature about this topic. We believe that this work deserves to be tested in long-term follow-up studies to determine its appropriate place in sepsis care in different regions of the world. Useful references include:

• Liberati A et al.Cochrane Collaboration. 2010;9:1–72

• de Jonge E et al. Lancet.2003;362:1011–1016

• de Smet et al. N Engl J Med.2009;360:20–31

• Cuthbertson BH et al. Trials.2010;11:117-120

• de Smet AMGA et al.Lancet Infect Dis. 2011;11:372–380

• Oostdijk EAN et al. Am J Respir Crit Care Med.2010;181:452–457

• Ochoa-Ardila MEet al. Intensive Care Med.2011;37:1458–1465

Routine infection control practices in the ICU are important, and recent guidelines on this topic have been published (Aitken et al. Crit Care Med.2011;39:1800-1818). Despite rather convincing and detailed meta-analysis of the value of SDD, SOD and oral rises with CHG (see SSC supplemental digital file 3), the practice remains variably applied and is a subject of ongoing controversy. oral CHG is often preferred based upon its ease of administration, safety, and the relative lack of concern over promoting antibiotic resistance within ICUs that widely use this practice. The existing evidence indicates that the use of SDD does not appear to be a major factor in promoting resistance in the ICU microbial flora. Three recent studies have studied this question. One suggested it does increase resistance to therapeutic antibacterials, one found no evidence of increased resistance with the use of SDD, and a third study found that the frequency of antibacterial resistance was actually reduced after the introduction of widespread use of SDD:

• de Smet AM et al. Lancet Infect Dis.2011;11:372-380

• Oostdijk et al. Am J Respir Crit Care Med.2010;181:452-457

• Ochoa-Ardila et al. Intensive Care Med2011;37:1457-1465

Clearly this question needs to be answered with additional long-term studies to verify the safety and efficacy of this method of infection prevention. Until such time as this information becomes available, a weak level 2 recommendation is given in support of SDD or its variants in regions with experience in its use.

新英格兰医学杂志全球投票结果

参加投票人数：全球共3221人

答案

ICU中去世的多数患者并非死于突发的意外并发症。多数死亡是可以预测的，死亡的确切时间取决于限制或撤除特殊医疗措施的决定。有关限制或撤除生命支持治疗措施的决策在国际上存在很大差异。其中一种极端的作法是，医生与家属分享相关信息，但在决策过程中居主导地位。另一种极端的作法则主张患者的自主权，医生仅仅发挥提出建议的作用。在北美、部分欧洲国家以及澳大利亚，常见的情况是医患双方共同参与决策过程。某些患者可能已经准备好有关延续生命治疗措施的生前预嘱；医患双方应当尊重这一预嘱，在部分地区甚至具有法律效力。无论决策过程如何，均应以确保及尊重患者意愿为目的。当预计患者即将死亡而限制或撤除治疗时，治疗目的应当根据个体情况，使其符合患者躯体、精神及文化需求。

我们的患者并无生前预嘱，但家属一致认为患者自己并不希望采取延续生命的支持治疗措施，除非能够恢复到既往的功能状态。在ICU长时间住院期间，家属与ICU医务人员建立了相互信任的关系，家属认为进行采取积极治疗措施并不符合患者意愿。在接下来的24小时内，希望探视患者的家属得到了最后探望患者的机会，此后拔除了气管插管，应用吗啡以避免出现呼吸困难。最终，在ICU的一个隔离房间中，在亲属的陪伴下，患者安详地去世。

相关文章

如欲了解更多内容，请参见新英格兰医学杂志2014年6月26日发表的危重病医学系列文章之结局篇，由D. Cook和G. Rocker撰写的Dying with Dignity in the Intensive Care Unit