Bloodstream Infections Among Patients Treated with Intravenous Epoprostenol or Intravenous Treprostinil for Pulmonary Arterial Hypertension --- Seven Sites, United States, 2003--2006

Pulmonary arterial hypertension (PAH) is a life-threatening disorder characterized by elevated pulmonary artery pressure and pulmonary vascular resistance. Continuous infusion of a prostanoid, which acts as a vasodilator and anti-proliferative agent, is indicated in the treatment of patients with severe PAH. Two prostanoids are approved for intravenous (IV) use in the United States: epoprostenol (epoprostenol sodium [brand name Flolan^®], Gilead, Foster City, California) and treprostinil (treprostinil sodium [brand name Remodulin^®], United Therapeutics, Silver Spring, Maryland) (1). These drugs are administered to PAH patients at hundreds of treatment centers in the United States. In September 2006, CDC received a report from a PAH specialist of a suspected increase in the number of gram-negative bloodstream infections (BSIs) among PAH patients treated with IV treprostinil. CDC conducted a retrospective investigation with the assistance of several state health departments and the cooperation of seven PAH treatment centers to determine the relative rates of BSI in a sample of patients treated with IV treprostinil and IV epoprostenol during 2003--2006. This report describes the results of that investigation, which indicated that, based on combined data from seven separate PAH treatment centers, pooled mean rates of BSI (primarily gram-negative BSI) were significantly higher for patients on treprostinil than for those on epoprostenol. The results do not suggest intrinsic contamination of IV treprostinil as a cause of the infections; the difference in rates might have been caused by differences in preparation and storage of the two agents, differences in catheter care practices, or differences in the anti-inflammatory activity of the agents. Health-care providers who care for PAH patients should be aware of these findings. Further investigation is needed to determine the causes of the different infection rates at centers where this was observed and to determine whether such a difference exists in other PAH treatment centers.

A BSI was defined as any positive blood culture result for a patient receiving IV epoprostenol or IV treprostinil as reported from a clinical laboratory or by a health-care provider in the medical record during the period for which data were collected. BSI rate information among patients on either IV prostanoid was requested from all PAH treatment centers that were major prescribers of IV treprostinil during 2006, as reported to investigators by United Therapeutics. BSI rates were calculated on the basis of the number of BSIs per 1,000 medicine days (i.e., days on which an IV prostanoid was administered). Because of varying arrangements between PAH treatment centers and investigators, the period and method of data collection varied by center. Data for calculating infection rates were available from seven different centers. Data for both agents were available from five centers; three reviewed their records to determine the numbers of infections and medicine days (one center retrospectively determined its rates for 2003--2006, one center for 2004--2006, and one center for 2006 only), and two invited state health departments and CDC to perform site visits to collect this information (data collected for 2004--2006). Data on IV treprostinil only was available from two centers, which provided CDC the number of BSIs among patients on IV treprostinil from January 1 to October 1, 2006; rates for these two centers were then calculated using medicine-day data specific to each center provided by United Therapeutics. Rates of BSI for patients receiving IV epoprostenol at these two centers could not be calculated because the total number of medicine days was not available.

A total of 51,183 medicine days were reported from the seven centers for IV treprostinil and 201,158 from five centers for IV epoprostenol. A total of 144 BSIs were identified (57 among patients on IV treprostinil and 87 among patients on IV epoprostenol). Overall, 26 different organisms (12 gram positive, 12 gram negative, and two acid fast bacilli) were identified in patients with BSIs. Fourteen different organisms were found in BSIs among patients on IV epoprostenol (nine gram positive, four gram negative, and one acid fast bacillus), and 18 different organisms were found in BSIs among patients on IV treprostinil (seven gram positive, 10 gram negative, and one acid fast bacillus). The three most common organisms isolated in BSIs among patients on IV epoprostenol were coagulase-negative Staphylococcus (28), Micrococcus sp. (18), and Staphylococcus aureus (12). The three most common organisms isolated in BSIs among patients on IV treprostinil were Pseudomonas sp. (11), Enterobacter sp. (nine), and coagulase-negative Staphylococcus (eight). The overall BSI pooled mean rate (per 1,000 medicine days) was higher for patients receiving IV treprostinil compared with those receiving IV epoprostenol (1.11 versus 0.43; pooled incidence rate ratio [IRR] = 2.57; 95% confidence interval [CI] = 1.81--3.64). Individual center BSI rates for treprostinil ranged from 0.28 to 2.10, and for epoprostenol from 0.23 to 1.02; IRR at individual centers ranged from 0.59 to 3.90. The pooled mean rate for gram-negative BSIs was higher among patients on IV treprostinil than on IV epoprostenol (0.76 versus 0.06; pooled IRR = 12.77; CI = 6.55--26.80). Individual center gram-negative BSI rates for treprostinil ranged from 0.28 to 1.11, and for epoprostenol from 0.03 to 0.25; IRR ranged from 1.19 to 29.90. Among the five centers for which rates for both drugs were available, overall BSI rates for treprostinil were higher than epoprostenol at four centers, and gram-negative BSI rates for treprostinil were higher at all five. Mantel-Haenszel--adjusted IRRs were calculated from the five centers that supplied rate information for both medications: 2.62 (CI = 1.84--3.72) for overall BSI and 13.53 (CI 6.72--27.22) for gram-negative BSI. Overall and gram-negative BSI rates for both patients receiving IV treprostinil and IV epoprostenol were higher than historic BSI rates among patients receiving IV epoprostenol (historic BSI rates: 0.15 per 1,000 medicine days; historic gram-negative BSI rates: 0.01 per 1,000 medicine days), as reported in a previously published, two-center study (2).

Pooled mean rates of overall and gram-negative BSIs for treprostinil remained higher than those for epoprostenol after stratification for several factors, including 1) method of data collection (data available from seven centers), 2) time receiving IV prostanoids (<2 years versus >2 years) (two centers), 3) age (<18 years versus >18 years) (two centers ), 4) home health-care company providing the medications and coordinating their administration (two centers), and 5) co-administration of immunosuppressive medications (two centers). BSI-related mortality was evaluated at three centers (the two centers that CDC investigators visited and for which this variable could be evaluated, and one center for which these data were available). Two deaths from two different centers were identified as possibly related to a BSI, both in patients who received IV treprostinil. One death was in a patient with a gram-negative BSI, and one was in a patient with a gram-positive BSI.

Ten partially used vials of IV treprostinil from BSI patients in five states were cultured for the presence of bacteria at the New York State Department of Health's Wadsworth Center laboratory, the New York City Department of Health laboratory, or CDC. No organisms were recovered from any of these vials.

Reported by: R Barst, MD, E Berman-Rosenzweig, MD, E Horn, MD, L Saiman, MD, Columbia Univ College of Physicians and Surgeons, New York; L Lee, MSc, New York City Dept of Health and Mental Hygiene; K Southwick, MD, D Kohlerschmidt, P Smith, MD, New York State Dept of Health. M Gomberg-Maitland, MD, S Weber, MD, Univ of Chicago Hospital, Chicago, Illinois. D Ivy, MD, A Doran, Children's Hospital, Univ of Colorado Health Sciences Center, Denver, Colorado. M McGoon, MD, C Anderson Severson, Mayo Clinic College of Medicine, Rochester, Minnesota. M Rubenfire, MD, V McLaughlin, MD, Univ of Michigan, Ann Arbor, Michigan. J Feldman, MD, C Tanner, Pulmonary Hypertension and Advanced Lung Disease Program, Phoenix, Arizona. K Chin, MD, M Kingman, Univ of Texas Southwestern Medical Center, Dallas, Texas. V Tapson, MD, Duke Univ Medical Center, Durham, North Carolina. A Srinivasan, MD, J Noble-Wang, PhD, M Arduino, DrPH, R Shoulson, E Petersen, National Center for Preparedness, Detection, and Control of Infectious Diseases (proposed); A Kallen, MD, E Lederman, MD, A Balaji, PhD, I Trevino, DVM, EIS officers, CDC.

Editorial Note:

Epoprostenol and treprostinil are prostanoids used in the treatment of severe PAH via continuous infusion. Epoprostenol, also called prostacyclin, was approved by the Food and Drug Administration (FDA) in 1995 and is the most frequently prescribed prostanoid. Patients must reconstitute the medication at home and keep it continually cool once mixed. It has a half-life of approximately 3--6 minutes, and sudden interruptions in the infusion can be life-threatening (1,3). Treprostinil was approved by FDA in 2004 for continuous IV infusion. This agent has several advantages over epoprostenol, including a longer half-life (approximately 4 hours) and the ability to remain stable without refrigeration (1). Both medications can be administered via a central venous catheter, putting patients at risk for catheter-related BSI. The medications are administered to patients primarily at home, and patients and their caregivers are primarily responsible for preparing and infusing medications and for changing dressings and IV tubing.

The investigation described in this report identified higher overall and gram-negative BSI rates among PAH patients receiving IV treprostinil compared with IV epoprostenol among patients treated at seven centers. Although the cause of this difference is unclear, the variety of organisms involved and the absence of bacterial growth in vials of used IV treprostinil obtained from patients with BSIs argue against intrinsic contamination of IV treprostinil or its diluents as a cause of the apparent BSI rate increase. At least three hypotheses exist that might explain the difference.

First, differences in practices involved in the preparation and storage of the two agents might create different levels of risk for BSI. IV epoprostenol is supplied as a powder that is mixed by the patient with single-use vials of a diluent, accessed by a single-use needle and syringe. IV treprostinil is supplied in a multiple-dose vial that the patient may use for up to 30 days, accessed either by a single-use needle and syringe or via a vial adaptor that may remain in place for as long as the vial is in use. Multiple-dose vials of medications have been implicated repeatedly in infection outbreaks in health-care settings, particularly when health-care providers used suboptimal infection-control practices or did not follow manufacturer-recommended storage conditions (4,5).

Second, differences in infection-control practices involved in central venous catheter and infusion-set care might create different levels of risks for BSIs. During the investigation, concerns were raised by several PAH specialists about infection-control practices associated with administration of these medications, including differences in line tubing and dressing changes, techniques for accessing vials of IV treprostinil, use of in-line filters, and the use of vial adaptors, which may be left in place for extended periods of time. In addition, the longer half-life and simpler administration requirements of treprostinil might lead to greater complacency regarding catheter care and use. For example, patients might disconnect the infusion for certain periods, thereby increasing the risk for BSI because breaks in the infusion system can allow for introduction of bacteria.

Third, the difference might be attributed to differences in the anti-inflammatory activity of the agents. However, studies of this subject in prostanoid analogues are limited because they all have been conducted in vitro, have not evaluated epoprostenol, and have used concentrations of prostacyclin analogues that are generally higher than those achieved in vivo for treatment of PAH (6--8).

The findings in this report are subject to at least four limitations. First, to maximize data collection for both medications, data were collected from several clinical centers that might not have been able to provide data for the entire period requested. Thus, the analyses were limited because 1) data were collected during different periods and were not available for both medications at all seven centers and 2) detailed information on potential risk factors for BSI were not available from most centers. Second, the sample of centers was not random, and the findings are subject to ascertainment bias because centers that noticed differences in rates might have been more likely to submit data; hence, the findings might not be representative of all PAH centers. Third, to be consistent across all data sources, a microbiologic definition of BSI was used rather than a clinical definition. Therefore, infections diagnosed on clinical grounds alone (i.e., without cultures) were not included, whereas positive blood cultures of unknown clinical significance (e.g., those not associated with clinical symptoms) might have been included. Finally, differences in BSI detection rates might have resulted from clinicians being more likely to obtain blood cultures from patients treated with one agent versus the other.

In the sample described in this report, the rate of BSI and gram-negative BSI in patients on IV treprostinil were higher than the rates in patients on IV epoprostenol. The rates reported in this report equate to one BSI every 2.5 years for a patient receiving IV treprostinil and every 6.4 years for a patient receiving IV epoprostenol. Further investigation is needed to identify the causes of the differences identified and to determine if these results are found at other PAH centers. Until further information is available, clinicians who administer these drugs should be aware of the potential differences in BSI risk, particularly regarding infections caused by gram-negative organisms, which can be difficult to treat and can lead to substantial morbidity and mortality. Because IV treprostinil offers certain advantages over IV epoprostenol, clinicians should advise patients about the potential advantages and disadvantages of both drugs and assist patients in choosing the medication that best suits their clinical situation. Clinicians should also be aware of the risk for gram-negative pathogens in patients receiving IV prostanoid infusions when choosing empiric antibiotic treatment for patients with possible catheter-associated infections. Health-care providers also should routinely review with patients the infection-control practices for the management of central venous catheters and administration of these medications.

Acknowledgment

The findings in this report are based, in part, on contributions by K Kaye, MD, S Evans, Y Choi, K Barritt, Duke Univ Medical Center, Durham; J Engel, MD, North Carolina Div of Public Health.

References

1. Lee SH, Rubin LJ. Current strategies for pulmonary artery hypertension. J Intern Med 2005;258:199--215.
2. Oudiz RJ, Widlitz A, Beckmann J, et al. Micrococcus-associated central venous catheter infections in patients with pulmonary artery hypertension. Chest 2004;126:90--4.
3. Barst RJ, Rubin LJ, Long WA, et al. A comparison of continuous intravenous epoprostenol (prostacyclin) with conventional therapy for primary pulmonary hypertension. N Eng J Med 1996;334:296--301.
4. Kirschke DL, Jones TF, Stratton CW, Barnett JA, Schaffner W. Outbreak of joint and soft tissue infections associated with injections from a multiple-dose medication. Clin Infect Dis 2003;36:1369--73.
5. Mattner F, Gastmeier P. Bacterial contamination of multiple-dose vials: a prevalence study. Am J Infect Control 2004;32:12--6.
6. Eisenhut T, Sinha B, Grottrup-Wolfers E, et al. Prostacyclin analogs suppress the synthesis of tumor necrosis factor-alpha in LPS-stimulated human peripheral blood mononuclear cells. Immunopharmacology 1993;26:259--64.
7. Lindemann S, Gierer C, Darius H. Prostacyclin inhibits adhesion of polymorphonuclear leukocytes to human vascular endothelial cells due to adhesion molecule independent regulatory mechanisms. Basic Res Cardiol 2003;98:8--15.
8. Aronoff DM, Peres CM, Serezani CH, et al. Synthetic prostacyclin analogs differentially regulate macrophage function via distinct analog-receptor binding specificities. J Immunol 2007;178:1628--34.

Use of trade names and commercial sources is for identification only and does not imply endorsement by the U.S. Department of Health and Human Services. References to non-CDC sites on the Internet are provided as a service to MMWR readers and do not constitute or imply endorsement of these organizations or their programs by CDC or the U.S. Department of Health and Human Services. CDC is not responsible for the content of pages found at these sites. URL addresses listed in MMWR were current as of the date of publication.

Disclaimer   All MMWR HTML versions of articles are electronic conversions from ASCII text into HTML. This conversion may have resulted in character translation or format errors in the HTML version. Users should not rely on this HTML document, but are referred to the electronic PDF version and/or the original MMWR paper copy for the official text, figures, and tables. An original paper copy of this issue can be obtained from the Superintendent of Documents, U.S. Government Printing Office (GPO), Washington, DC 20402-9371; telephone: (202) 512-1800. Contact GPO for current prices.

**Questions or messages regarding errors in formatting should be addressed to [email protected].

Date last reviewed: 3/1/2007