A 75 year old woman is found to have pneumonia. With a CURB-65 of 3 and a PORT score of 95, she is correctly treated her for community acquired pneumonia with Ceftriaxone and Azithromycin, and admitted. Unfortunately, the admitting service points out that no blood cultures were drawn! What is the evidence for this?
Originally from Clinical Monster blog
Official Guidelines and Recommendations
I have been told that CMS
UPDATE 2/17/14: As of January 1, 2014, the CMS and JCAHO are retiring blood cultures for non-ICU patients as a core measure (PN-3b measure). Also the Surviving Sepsis Campaign recommendations are for pneumonia patients with severe sepsis only. Thanks to Drs. Seth Trueger, Scott Weingart, and @DocERTrauma for this correction.
What is the evidence for this?
Two studies retrospectively reviewed total blood culture yield. One  had 1.4% true positive blood cultures with only 0.18% affecting patient management. The other  yielded 5% true positives with 1.6% (18 patients) affecting patient management. By combing these studies we find a 0.7% chance of affecting patient care. That’s a NNT of 143! Additionally, patients with false positive cultures had significantly longer length of stays leading to increased cost.
Benenson RS et al. J Emerg Med (2007)
This retrospective review  examined 684 patient charts with ED blood cultures and a discharge diagnosis of pneumonia. They found 23 (3.4%) true positive and 54 (7.8%) false positives. Of the true positives, 3 had their antibiotic regimen narrowed without anyone needing broader coverage. That’s a 0.4% chance of a blood culture drawn affecting patient management, or an NNT of 250! Interestingly, 18 additional patients could have been narrowed based on culture results, but were not. The authors recommended “eliminating blood cultures for CAP patients and obtaining blood cultures for HCAP patients presenting to the ED”
Kennedy M et al. Ann Emerg Med (2005)
This retrospective review  demonstrated that of 414 ED blood cultures drawn for pneumonia, 29 (7%) were true positives and 25 (6%) were false positives/contaminants. Of the 29 true positives, 11 (2.7%) had their coverage narrowed, 4 (1%) had their coverage broadened, and another 8 could have been narrowed based on culture results but were not. In this study, the rate of blood culture results affecting patient care was higher at 3.6%. Of note, 3 of the 4 that needed broader coverage were from a nursing home, had MRSA, but had not been initially treated with hospital acquired pneumonia regimens.
Coburn B et al. JAMA (2012)
This literature search/meta-analysis  investigated the risk of true positive blood cultures in immune-competent patients without suspicion for endocarditis. They did recommended NOT to do a blood culture in the Low Risk Group due to the low pre-test probability of a true positive culture! Below is a summary of their findings:
|Risk Group||Pretest Probability of Positive Blood Culture||Recommend Blood Culture?||Diagnoses Included|
|High||38-68%||Yes||Severe sepsis, septic shock, bacterial meningitis|
In terms of risk factors for a true positive culture, the following were NOT statistically significant:
- Subjective fever
- Tachycardia alone
- Elevated WBC count
- Documented fever
The following WERE statistically significant:
- Shaking chills
- Requiring vasopressors
- Neutrophil/lymphocyte ratio > 10
- Presence of SIRS
Interestingly, having zero SIRS criteria had a LR 0.09 for a true positive culture.
Shapiro NI et al. J Emerg Med (2008)
This prospective analysis included 3,730 pneumonia patients with blood cultures drawn in the ED or up to 3 hours after admission. In the derivation population, they found a statistically significant increased risk of positive blood culture with certain characteristics. After some moderately complicated calculations, they created Major Criteria and Minor Criteria and recommended a blood culture only if the patient has one major or two minor criteria:
|Major criteria||Minor criteria|
|• Suspected endocarditis|
• Temperature > 103F
• Indwelling vascular catheter
|• Temperature > 101F
• Age > 65 years
• Systolic BP < 90 mmHg
• WBC > 18k
• Bands > 5%
• Platelets < 150,000
• Creatinine > 2 mg/dL
This approach had a negative predictive value for a true positive blood culture of 99.4% (95% CI 99-100%) and 99.1% (95% CI 98-100%) in the derivation group and validation group, respectively. Seven total patients were missed by the decision rule. Of those, five would have had no change in management, one should have received a culture by the prediction rule but was missed in the ED, and one received a blood culture for a fever > 3 hours after admission.
So what do these organizations actually say?
IDSA and ATS guidelines for Community Acquired Pneumonia (CAP)
These organizations recommend blood cultures for patients with CAP only if they have at least one for the following:
- Admission to the intensive care unit
- Cavitary infiltrates
- Chronic severe liver disease
- Plural effusion
- A positive pneumococcal urinary antigen test
- Active alcohol abuse
Bonus Clinical Pearls
- If cultures are performed, the literature emphasizes that volume matters . There is a 3% increased yield in positive blood cultures per mL of blood obtained. They recommend at least 7 mL per blood culture bottle.
- Additionally, the optimal time to culture a patient is 1-2 hours prior to fever occurring . Multiple studies have shown no significant difference in results if the patient is cultured at Tmax vs hours before vs hours after. According to this data, you don’t need to rush to do cultures when someone is febrile.
- Another study  showed that contaminated blood culture rates increased linearly with increased patient load in the ED.
- Obtaining blood cultures for non-ICU CAP are NO LONGER core measures per CMS and JCAHO as of January 1, 2014, finally in alignment with studies showing their low utility value in low risk patients. [Updated 2/17/14]
- Blood cultures are not universally required for admitted CAP patients per IDSA/ATS, but required per CMS/JCAHO for ICU patients (PN-3a measure). [Updated 2/18/14]
- Cultures rarely affect management, but are recommended by the literature when certain criteria are met including sepsis and endocarditis.
- If blood cultures are obtained, clean well, don’t hurry, don’t worry wait until patient is febrile, and fill up bottles with at least 7 mL each.
- Overview of Specifications of Measures Displayed on Hospital (pdf)
- Mandell LA, et al; Infectious Diseases Society of America; American Thoracic Society. Infectious DiseasesSociety of America/American Thoracic Society consensus guidelines on the management of community-acquired pneumonia in adults. Clin Infect Dis. 2007 Mar 1;44 Suppl 2:S27-72. PMID: 17278083 (pdf)
- Joint Commission’s Pneumonia National Hospital Inpatient Quality Measures (pdf)
- Surviving Sepsis Campaign Initial Resuscitation (pdf)
- Howie N, Gerstenmaier JF, Munro PT. Do peripheral blood cultures taken in the emergency department influence clinical management? Emerg Med J. 2007 Mar;24(3):213-4. PMID: 17351231
- Kelly AM. Clinical impact of blood cultures taken in the emergency department. J Accid Emerg Med. 1998 Jul;15(4):254-6. PMID: 9681310
- Benenson RS, Kepner AM, Pyle DN 2nd, Cavanaugh S. Selective use of blood cultures in emergency department pneumonia patients. J Emerg Med. 2007 Jul;33(1):1-8. PMID 17630066
- Kennedy M, Bates DW, Wright SB, Ruiz R, Wolfe RE, Shapiro NI. Do emergency department blood cultures change practice in patients with pneumonia? Ann Emerg Med. 2005 Nov;46(5):393-400. PMID 16271664
- Coburn B, Morris AM, Tomlinson G, Detsky AS. Does this adult patient with suspected bacteremia require blood cultures? JAMA. 2012 Aug 1;308(5):502-11. PMID 2285117
- Shapiro NI, Wolfe RE, Wright SB, Moore R, Bates DW. Who needs a blood culture? A prospectively derived and validated prediction rule. J Emerg Med. 2008 Oct;35(3):255-64. PMID: 18486413.
- Mermel LA, Maki DG. Detection of bacteremia in adults: consequences of culturing an inadequate volume of blood. Ann Intern Med. 1993 Aug 15;119(4):270-2. PMID: 8328734.
- Riedel S, et al. Timing of specimen collection for blood cultures from febrile patients with bacteremia. J Clin Microbiol. 2008 Apr;46(4):1381-5. doi: 10.1128/JCM.02033-07. Epub 2008 Feb 27. Erratum in: J Clin Microbiol. 2008 Jul;46(7):2475. PMID:
- Halverson S, Malani PN, Newton DW, Habicht A, Vander Have K, Younger JG. Impact of hourly emergency department patient volume on blood culture contamination and diagnostic yield. J Clin Microbiol. 2013 Jun;51(6):1721-6. PMID: 23515549