1. Home>
2. Specialties>
3. Emergency Medicine>
4. Letter to Readers

The CMS Blood Cultures for CAP Program: The Architects Speak Out

An exchange of views on the use of blood cultures in managing community-acquired pneumonia

Response to "Blood cultures aren't useful for managing immunocompetent CAP inpatients"

To the Editor: In the November 24, 2004, issue of Journal Watch Emergency Medicine, Birnbaumer¹ summarized the recent publication of Corbo et al. on use of initial blood cultures in community-acquired pneumonia (CAP).² Those researchers found that of 355 immunocompetent, community-dwelling adults who were admitted because of CAP and who had blood cultures drawn, 9% had true-positive cultures and 10% had false-positive results. Antibiotics were changed in 76% of cases with true-positive results, with 10 of 25 (40%) changes attributed to culture results, and in 70% of cases with false-positive results, with 6 of 26 (23%) changes attributable to culture results.

Birnbaumer comments that this and previous studies show blood cultures to have "limited or no value" in guiding empirical antibiotic treatment of immunocompetent adults with CAP. She suggests that the Centers for Medicare and Medicaid Services' (CMS's) clinical performance measure that blood cultures be performed before administering antibiotics for all hospitalized pneumonia patients is an attempt to ". . . impose its own definition of ‘quality.' " Finally, she suggests that "maybe CMS decision makers should take a peek at the literature." As leaders of or contributors to the CMS National Pneumonia Project (NPP), we welcome the opportunity to respond to Dr. Birnbaumer's comments.

Pneumonia accounts for the acute-care hospitalization of more than 600,000 Medicare beneficiaries each year and is associated with a 30-day mortality rate of about 12%. Because of pneumonia's impact, CMS uses several pneumonia inpatient-care performance measures in the NPP. Among them are (since 1999) collection of blood cultures before the initial administration of antibiotics and (since 2002) performance of cultures within 24 hours of admission. However, these measures do not constitute an attempt by CMS to "impose its own definition of quality," as Birnbaumer suggests. Rather, CMS uses these measures because the American Thoracic Society (ATS), the Canadian Infectious Diseases Society, the Canadian Thoracic Society, and the Infectious Diseases Society of America (IDSA) have included in their official recommendations through 2004 two sets of blood cultures for all adult CAP inpatients.^3,4,5,6,7,8 Collection of blood samples before administration of antibiotics is recommended because prior treatment results in a reduction in yield of pathogens by at least one half.^9,10,11 In addition to publishing these recommendations, the societies have provided direct guidance to the NPP since its development in 1998.

Should the routine collection of blood cultures from all CAP patients be questioned? Yes, it should, just as any "routine" practice should be reconsidered periodically. Are the "CMS decision makers" aware that this practice has been questioned? Yes, they are. In addition to reviewing the pneumonia literature regularly, they have made contributions to it. Recent contributions include reports that used NPP data to identify subsets of CAP patients who are especially likely to have bacteremia⁹ and to describe clinicians' failure to narrow antibiotic coverage in response to culture results.¹²

As part of their periodic revision of CAP guidelines, the ATS and IDSA are reviewing all available evidence related to blood cultures, including that from the NPP.⁹ They will consider all of the issues raised in the literature, but also the fact that blood culture is often the only microbiologic test done in CAP evaluation, making it the primary source of the sensitivity data that are necessary to maintain the value of empiric treatment recommendations. It is premature to predict what the panel will recommend. Options under consideration include limiting the recommendation for blood cultures to more severely ill patients and basing the number of recommended cultures on clinical characteristics. These revised guidelines are expected to be published in 2005. The ongoing evolution of Medicare NPP performance measures will be guided by the societies' recommendations.

Peter M. Houck, MD, Centers for Medicare and Medicaid Services, Seattle

Dale W. Bratzler, DO, MPH, Oklahoma Foundation for Medical Quality, Oklahoma City

John G. Bartlett, MD, Johns Hopkins University School of Medicine, Baltimore

Lionel Mandell, MD, McMaster University, Hamilton, Ontario

Michael S. Niederman, MD, Winthrop University Hospital, Mineola, NY

The JWEM editors' reply

We thank the architects of the Centers for Medicare and Medicaid Services (CMS) Community Acquired Pneumonia (CAP) Quality Initiatives for their letter describing the rationale for the requirement that blood cultures be performed prior to antibiotic administration. The study in question, although small, adds another brick to a wall of data showing that such cultures have essentially no clinical value.^1,2 We appreciate the thoughtful letter by Houck and his colleagues and support the overall efforts by CMS to improve patient care. However, unlike prior CMS initiatives, such as use of aspirin in acute myocardial infarction, the CAP blood culture measure is not supported by evidence that it improves outcomes.

The CMS CAP initiative places enormous pressure on emergency departments to obtain blood cultures and to rapidly (within 4 hours of arrival) administer the first dose of an antimicrobial agent, even if this disrupts the system of care for the rest of the ED patient population. EDs are already dangerously overcrowded and bear the major burden of providing care to the nation's more than 40 million uninsured people by unfunded federal mandate. Now EDs must also meet the CAP performance targets, or they risk being identified as providing inferior care via "report cards" posted on the internet. In this context, it is mystifying to us why CMS chose to prioritize CAP as an area of focus and to impose these new requirements, given the absence of high-quality, outcome-based studies demonstrating that such measures benefit CAP patients.

In their letter defending the blood culture measure, the authors cite only consensus statements issued by the American Thoracic Society (ATS), the Infectious Diseases Society of America (IDSA), the Canadian Infectious Diseases Society, and the Canadian Thoracic Society. Why would these authors, and, by extension, CMS, choose to rely exclusively on consensus statements when there is a wealth of primary research available? And why would they base their recommendations on consensus statements when results from many studies unequivocally demonstrate that indiscriminate use of blood cultures in CAP is clinically worthless, expensive, and potentially harmful? One possible explanation lies in the fact that the authors of the letter, who represent themselves as the primary architects of the CMS CAP measures, are also the primary authors of these very same consensus statements on which the measures are based.^3,4,5,7,8 This is particularly vexing to those of us in academic centers who, for years, have analyzed the very same literature and used its uniform findings as the basis for training house staff to abandon blood cultures in CAP because they are not indicated and are wasteful.

Blood cultures drawn for CAP rarely identify a causative organism. In an undifferentiated population of patients admitted to the hospital with CAP, the rate of true-positive (noncontaminant) blood cultures ranges from 5.7%¹³ to 9.0%.¹ Rates of true-positive blood cultures for specific CAP populations range from as low as 0% in patients with nonsevere CAP¹⁴ and 2.1% in CAP patients discharged from the ED,¹⁵ to only 19.6% in the most critically ill, intubated, intensive care unit patients with CAP.¹⁶ More recent evidence suggests that there may be no correlation between severity of illness and likelihood of positive cultures.¹³

Even when blood cultures are positive, they rarely have any meaningful effect on clinical management or antibiotic choice. Multiple studies have consistently shown that antibiotic coverage is narrowed in only 20%-30% of the small number of patients with positive cultures. Coverage is broadened in only 3%-5% of patients with positive cultures, representing 0.1%-0.3% of the overall admitted CAP population.^1,13,17,18 In 58% of patients with positive cultures, the original antibiotic course is continued despite culture results indicating that coverage can be narrowed. Blood culture results also have no effect on use of fluoroquinolones in the inpatient setting.¹² For the vast majority of cases in which antibiotic coverage is changed, the change is prompted by clinical response, regardless of culture results.^1,19

Ample data show no association between obtaining blood cultures before antibiotic administration and improved clinical outcome. A study of 1062 admitted CAP patients showed that obtaining blood cultures either prior to antibiotic administration or within 24 hours of admission was not associated with any of three key quality indicators: inpatient mortality, clinical stability at 48 hours, and shorter length of stay.²⁰ Even when the population was restricted to the elderly, a retrospective study of 14,000 patients failed to show any significant association between obtaining blood cultures prior to antibiotic administration or within 24 hours of admission and all-cause 30-day mortality.²¹ Paradoxically, this study is the one most frequently cited to justify universal CAP blood cultures, apparently because the statistically nonsignificant finding in this very large study was nearly significant (the adjusted odds ratio for 30-day mortality in patients who had blood cultures compared with those who did not have blood cultures was 0.90, with a statistically nonsignificant P value of 0.07). Furthermore, only 57% of patients in the study who were supposed to have cultures actually got them, indicating enormous selection bias and challenging the validity of any conclusions. Perhaps diligence in performing the cultures was correlated with overall quality and diligence of care.

Positive blood cultures also are not associated with any change in duration of treatment or with the success of switching otherwise clinically stable patients from intravenous to oral antibiotic treatment.²² In one study, mortality rates did not differ between patients whose antibiotic switch was guided by blood-culture results and those who were switched empirically based on clinical response (25% and 16%, respectively).²³

One major danger of indiscriminate blood cultures is a high false-positive rate. In an undifferentiated population, the false-positive rate approaches the true-positive rate; patients with false-positive results are as likely to have culture-guided antibiotic changes as patients with true-positive results^23,24. Of more concern, a 1991 study showed that patients with false-positive cultures had longer hospital stays (mean increase, 4.5 days) and greater charges (mean increase, $5674) compared with patients who had negative cultures.

The authors of the letter and architects of the CMS quality measures are prominent and respected leaders in pneumonia research, so they must be aware of the compelling data that clearly oppose the practice of obtaining blood cultures in all CAP patients. In the letter to JWEM, Houck and colleagues offer only one other justification for this recommendation, which, through the actions of CMS, has become a firm requirement: They argue that blood cultures are the "primary source of the sensitivity data that are necessary to maintain the value of empiric treatment recommendations."

Despite the extremely low yield of blood cultures in CAP and the fact that the most common organism implicated in CAP cannot be detected by culture, let us nevertheless assume that the above surveillance argument makes sense from a public health/epidemiology standpoint and that the scant surveillance data are of greater value than outcome studies comparing various empirical antimicrobial regimens. The authors, with their research interests in pneumonia, have worked within their respective societies to make the recommendation that blood cultures be obtained in patients with CAP. Subsequently, they authored guidelines for CMS that establish CAP blood cultures as a quality measure, in effect imposing an unfunded public-health research mandate on hospitals, insurers, and patients, with no support from evidence-based medicine or proven improvement in quality of care or outcomes. Ironically, the blood culture mandate confounds ED efforts to meet another quality standard within the same program -- namely, prompt administration of antibiotics.

But, does the surveillance argument make sense? The practice of using CAP blood cultures as the primary source of infection and resistance patterns is itself fraught with the potential to mislead. Microbiology data obtained exclusively from the 15%-20% of CAP patients admitted to the hospital (representing 0.5%-1.0% of the total population of pneumonia patients) cannot validly be considered to represent infection and resistance patterns in the community as a whole. Further, the required 4-hour window for administering antibiotics creates a very strong incentive to "shoot first and ask questions later" in order to meet the requirements, should the patient ultimately be admitted for CAP. This incentive promotes indiscriminate use of "inpatient-level" antibiotics (fluoroquinolones, third-generation cephalosporins), which may inevitably result in increased resistance to these vital drugs.

CMS must revisit this issue. The imposition of the CAP blood culture mandate on the nation's EDs was well intended, but ill conceived. To continue it despite awareness of the enormous financial, logistical, and quality-of-care burden that it has created is not appropriate.

Ron M. Walls, MD, FRCPC, FACEP, and Joshua B. Resnick, MD, MBA

Dr. Walls is Editor-in-Chief of Journal Watch Emergency Medicine. Dr. Resnick is an emergency medicine resident in the Brigham and Women's Hospital and Massachusetts General Hospital Harvard-affiliated Emergency Medicine Residency program in Boston.

Published in Journal Watch Emergency Medicine April 27, 2005

Citation(s):

1. Birnbaumer DM. Blood cultures aren't useful for managing immunocompetent CAP inpatients. JWEM Nov 24 2004.

2. Corbo J et al. Limited usefulness of initial blood cultures in community acquired pneumonia. Emerg Med J 2004 Jul; 21:446-8.

• Original article (Subscription may be required)
• Medline abstract (Free)

3. Niederman MS et al. Guidelines for the initial management of adults with community-acquired pneumonia: Diagnosis, assessment of severity, and initial antimicrobial therapy. American Thoracic Society. Medical Section of the American Lung Association. Am Rev Respir Dis 1993 Nov; 148:1418-26.

• Medline abstract (Free)

4. Niederman MS et al. Guidelines for the management of adults with community-acquired pneumonia. Diagnosis, assessment of severity, antimicrobial therapy, and prevention. Am J Respir Crit Care Med 2001 Jun; 163:1730-54.

• Original article (Subscription may be required)
• Medline abstract (Free)

5. Mandell LA et al. Canadian guidelines for the initial management of community-acquired pneumonia: An evidence-based update by the Canadian Infectious Diseases Society and the Canadian Thoracic Society. The Canadian Community-Acquired Pneumonia Working Group. Clin Infect Dis 2000 Aug; 31:383-421.

• Original article (Subscription may be required)
• Medline abstract (Free)

6. Bartlett JG et al. Community-acquired pneumonia in adults: Guidelines for management. The Infectious Diseases Society of America. Clin Infect Dis 1998 Apr; 26:811-38.

• Medline abstract (Free)

7. Bartlett JG et al. Practice guidelines for the management of community-acquired pneumonia in adults. Infectious Diseases Society of America. Clin Infect Dis 2000 Aug; 31:347-82.

• Original article (Subscription may be required)
• Medline abstract (Free)

8. Mandell LA et al. Update of practice guidelines for the management of community-acquired pneumonia in immunocompetent adults. Clin Infect Dis 2003 Dec 1; 37:1405-33.

• Original article (Subscription may be required)
• Medline abstract (Free)

9. Metersky ML et al. Predicting bacteremia in patients with community-acquired pneumonia. Am J Respir Crit Care Med 2004 Feb 1; 169:342-7.

• Original article (Subscription may be required)
• Medline abstract (Free)

10. Marston BJ et al. Incidence of community-acquired pneumonia requiring hospitalization. Results of a population-based active surveillance study in Ohio. The Community-Based Pneumonia Incidence Study Group. Arch Intern Med 1997Aug 11/25; 157:1709-18.

• Medline abstract (Free)

11. Porath A et al. The epidemiology of community-acquired pneumonia among hospitalized adults. J Infect 1997 Jan; 34:41-8.

• Medline abstract (Free)

12. Chang NN et al. Blood culture and susceptibility results and allergy history do not influence fluoroquinolone use in the treatment of community-acquired pneumonia. Pharmacotherapy 2005 Jan; 25:59-66.

• Medline abstract (Free)

13. Campbell SG et al. The contribution of blood cultures to the clinical management of adult patients admitted to the hospital with community-acquired pneumonia: A prospective observational study. Chest 2003 Apr; 123:1142-50.

• Medline abstract (Free)

14. Theerthakarai R et al. Nonvalue of the initial microbiological studies in the management of nonsevere community-acquired pneumonia. Chest 2001 Jan; 119:181-4.

• Medline abstract (Free)

15. Campbell SG et al. Utility of blood cultures in the management of adults with community acquired pneumonia discharged from the emergency department. Emerg Med J 2003 Nov; 20:521-3.

• Original article (Subscription may be required)
• Medline abstract (Free)

16. Rello J et al. Microbiological testing and outcome of patients with severe community-acquired pneumonia. Chest 2003 Jan; 123:174-80.

• Medline abstract (Free)

17. Waterer GW et al. The impact of blood cultures on antibiotic therapy in pneumococcal pneumonia. Chest 1999 Nov; 116:1278-81.

• Medline abstract (Free)

18. Chalasani NP et al. Clinical utility of blood cultures in adult patients with community-acquired pneumonia without defined underlying risks. Chest 1995 Oct; 108:932-6.

• Medline abstract (Free)

19. Sanyal S et al. Initial microbiologic studies did not affect outcome in adults hospitalized with community-acquired pneumonia. Am J Respir Crit Care Med 1999 Jul; 160:346-8.

• Original article (Subscription may be required)
• Medline abstract (Free)

20. Dedier J et al. Processes of care, illness severity, and outcomes in the management of community-acquired pneumonia at academic hospitals. Arch Intern Med 2001 Sep 24; 161:2099-104.

• Medline abstract (Free)

21. Meehan TP et al. Quality of care, process, and outcomes in elderly patients with pneumonia. JAMA 1997 Dec 17; 278:2080-4.

• Medline abstract (Free)

22. Ramirez JA and Bordon J. Early switch from intravenous to oral antibiotics in hospitalized patients with bacteremic community-acquired Streptococcus pneumoniae pneumonia. Arch Intern Med 2001 Mar 26; 161:848-50.

• Medline abstract (Free)

23. Waterer GW and Wunderink RG. The influence of the severity of community-acquired pneumonia on the usefulness of blood cultures. Respir Med 2001 Jan; 95:78-82.

• Medline abstract (Free)

24. Bates DW et al. Contaminant blood cultures and resource utilization: The true consequences of false-positive results. JAMA 1991 Jan 16; 265:365-9.

• Medline abstract (Free)

Image Gallery