News Release

Blood loss from lab testing associated with hospital-acquired anemia for patients with heart attacks

Peer-Reviewed Publication

JAMA Network

In patients with acute myocardial infarction (heart attack), blood loss from greater use of phlebotomy (blood drawn for diagnostic testing) appears to be independently associated with the development of hospital-acquired anemia (HAA), according to a report published Online First by Archives of Internal Medicine, one of the JAMA/Archives journals. The article is part of the journal's Less Is More series.

Anemia (low red blood cell count or low hemoglobin level) is associated with greater mortality and worse health status in patients with AMI, whether the condition is chronic (present at hospital admission) or hospital-acquired, according to background information in the article. However, HAA may be preventable if strategies to reduce blood loss in high-risk patients are implemented. Among the factors that may be associated with HAA is blood loss from diagnostic phlebotomy, which has been associated in other patient populations with in-hospital declines in hemoglobin level and need for blood transfusion. "Blood loss from phlebotomy could be an actionable target for intervention," write the authors.

Adam C. Salisbury, M.D., M.Sc., from Saint Luke's Mid America Heart and Vascular Institute, Kansas City, Mo., and colleagues analyzed data from the Cerner Corp.'s Health Facts database. Records selected from January 2000 through December 2008 included 17,676 patients who were admitted with AMI and no anemia from 57 hospitals. To measure blood loss, the researchers identified all phlebotomy events in patient records. They determined, by the laboratory tests ordered, which type of hematology tubes were used and the blood volume each held. Then, for every patient, the researchers multiplied these blood volumes by the number of tubes of each type that were collected during hospitalization. They also calculated the mean (average) blood drawn for every 24 hours of hospitalization and the mean phlebotomy volumes for each of the first 10 days of hospitalization.

About 20 percent of patients (n = 3,551) developed moderate to severe HAA. Estimates of the mean blood loss from phlebotomy for these patients (173.8 milliliters [mL]) were almost 100 mL higher than estimated blood loss in patients who did not develop moderate to severe HAA (83.5 mL) over the course of the hospitalization. The volume of diagnostic blood drawn was associated with developing moderate to severe HAA. The relative risk for developing HAA increased by 18 percent for each 50 mL of blood drawn; the relationship persisted when researchers adjusted the data for site and potential confounders. The average volume of blood drawn varied significantly across hospitals.

"In conclusion, blood loss from phlebotomy is substantial in patients with AMI, varies across hospitals, and is independently associated with the development of HAA," write the authors. "Studies are needed to test whether strategies that limit both the number of blood draws and the volume of blood removed for diagnostic testing can prevent HAA and improve clinical outcomes in patients with AMI."

(Arch Intern Med. Published online August 8, 2011. doi:10.1001/archinternmed.2011.361. Available pre-embargo to the media at

Editor's Note: Coauthors Drs. Salisbury, Spertus and Kosiborod are funded in part by an award from the American Heart Association Pharmaceutical Round Table and David and Stevie Spina. Please see the article for additional information, including other authors, author contributions and affiliations, financial disclosures, funding and support, etc.

Commentary: Hazards of Hospitalization—More Than Just 'Never Events'

An accompanying commentary by Stephanie Rennke, M.D., and Margaret C. Fang, M.D., M.P.H., both from the University of California, San Francisco, evaluated the study by Salisbury and colleagues in the larger context of patient safety in hospitals. "Of particular interest was the finding that the mean phlebotomy volume in patients varied widely across individual hospitals, suggesting that some blood tests may have simply been 'routine' and implying that reduction in the variability of care could potentially lead to reductions in HAA," Rennke and Fang write.

They cite increased attention to patient safety since the publication of an Institute of Medicine report on the topic in 1999. This focus, Rennke and Fang explain, "has led hospitals to test and implement system-wide interventions, some of which have been found to be effective in reducing the rates of hospital-acquired complications." As an example, they point to a Michigan Health and Hospital Association Keystone program which appeared to reduce rates of hospital-acquired infections and subsequent hospital deaths.

"With the increasing evidence that health care system interventions can reduce or prevent many hospital-acquired complications, efforts to implement effective strategies to make medical care safer and more effective are crucial," write the authors. "As Salisbury et al highlight in their study, HAA could potentially be considered a hazard of hospitalization. Investigations on how to modify this risk (e.g., through reducing unnecessary phlebotomy or reducing the volume of blood obtained during a hospitalization) could provide important insights into how to reduce anemia in the hospital and improve the value and appropriateness of care."


(Arch Intern Med. Published August 8, 2011. doi:10.1001/archinternmed.2011.335. Available pre-embargo to the media at

Editor's Note: Please see the article for additional information, including other authors, author contributions and affiliations, financial disclosures, funding and support, etc.

To contact corresponding author Mikhail Kosiborod, M.D., call Kim Cook at 816-932-3485 or e-mail To contact commentary coauthor Margaret C. Fang, M.D., M.P.H., call Karin Rush-Monroe at 415-502-NEWS or e-mail

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.