Entire newsletter text from the December 1, 2005 issue
Fatal misadministration of IV vincristine
As we began our Thanksgiving holiday last year, we had just learned that a 69-year-old woman had died due to a medication error that was caused in large part by an easily remedied safety problem-mistaken identity of products in unlabeled basins during a sterile procedure (ISMP Medication Safety Alert! Loud wake-up call: unlabeled containers lead to patient's death. December 2, 2004). This Thanksgiving holiday season, we were saddened and alarmed to learn that yet another easily remedied safety problem- accidental intrathecal administration of IV vincristine-had claimed the life of a 21-year-old man just a few months ago. A syringe containing vincristine for another patient had been accidentally delivered to the patient's bedside. A physician administered the drug intrathecally, believing it was a different medication intended for that route of administration. The patient died 3 days later.
Since 1989, ISMP has repeatedly published reports of these fatal errors. This latest fatality occurred just a month after the Joint Commission issued a Sentinel Event Alert (July 14, 2005), which also warned clinicians about this particular error and suggested ways to prevent its occurrence. The most recent alert was prompted by more than a decade of hauntingly similar tragic errors, which have continued despite a 1991 FDA and USP requirement to: (1) affix special warning labels (FATAL if given intrathecally, FOR IV USE ONLY) on extemporaneously prepared syringes, and (2) place syringes in manufacturer-supplied overwraps with a similar warning.
In our May 1, 2003 newsletter, which reported the death of a 49-year-old man after a similar error, we offered an error-reduction strategy that we had not previously published:
---Dilute vincristine in a plastic minibag of IV fluids for slow, continuously supervised (observed) infusion to deter confusion with intrathecal syringes. (Dilution stability has been fully supported by: Trissel LA, et al. The stability of diluted vincristine sulfate used as a deterrent to inadvertent intrathecal injection. Hospital Pharmacy 2001;36:740-745.)
This strategy ensures that IV vincristine will look distinctly different than a syringe containing an intrathecal medication. It also places the drug in a larger volume of fluid and in a different device for drug administration (infusion from a bag via IV tubing), neither of which lends itself well to intrathecal administration. While some of the root causes of errors associated with accidental intrathecal administration of IV vincristine have varied, all of these errors may have been avoided by this relatively simple error-reduction strategy (which the Joint Commission has also promoted in its recent Sentinel Event Alert).
Since our initial publication of this error-prevention strategy in 2003, some healthcare providers have successfully adopted its use. Others have voiced concern, citing the risk of extravasation as a significant deterrent, especially if IV vincristine is administered via a peripheral vein. Use of a peripheral vein may be fairly common, for example, in children who have had their central lines removed after the initial phase of chemotherapy, and are now receiving monthly treatments. Some healthcare practitioners, particularly nurses and physicians, have expressed that giving a vesicant into a peripheral vein warrants a smaller volume of solution so the nurse can continually validate venous access patency. And yet, diluted IV vincristine can be administered at a rate of infusion similar to the undiluted form, and the patient can be monitored in the exact same manner, even if vincristine has been diluted in a small volume of fluid. Thus, there has been some controversy regarding the safest way to administer the drug to reduce the risk of intrathecal administration without increasing the risk of harmful extravasations.
A recent, preliminary survey of 35 US children's hospitals from the Children's Oncology Group (COG) revealed that about one quarter (8, or 23%) are currently diluting IV vincristine, ostensibly without an increase in harm from extravasation. (Perhaps an unknown advantage of diluting vincristine is less tissue damage in the event that an extravasation does occur.) To learn how other institutions are administering IV vincristine, we are now repeating this survey and expanding it to include all pediatric and adult healthcare providers who administer IV and intrathecal chemotherapy. We are also including additional error-reduction strategies to assess the steps organizations have taken to prevent inadvertent intrathecal administration of IV vincristine. If you provide oncology services as specified, we would appreciate your help by completing a short survey on page 4.
After the initial results from this survey have been analyzed, we plan to gather additional responses from a small, representative sample of healthcare providers early in 2006 and will report our findings in this newsletter. With your help, we hope to raise the urgency with which organizations pursue eradication of this rare but fatal and preventable sentinel event.
Just before the pharmacy shift change at midnight, an order was received for the antibiotic daptomycin (CUBICIN) IV every 24 hours for an oncology patient who had a resistant gram-positive infection. The night pharmacist had never dispensed daptomycin, but she was aware of the drug and had asked an evening pharmacist to help her locate it in the pharmacy. The evening pharmacist obtained the daptomycin from the refrigerator and handed it to the night pharmacist. The night pharmacist typed "DA" into the pharmacy computer system, chose the drug from a list of medications and strengths on the screen, and generated an IV label. She then read the package insert, prepared the proper dose of daptomycin, affixed the label to the bag, and sent the drug to the unit where it was later administered. The next morning, a pharmacist was reviewing all antibiotic orders from the previous day and noted the handwritten daptomycin order. When he reviewed the patient's electronic drug profile, he could not find an entry for daptomycin. However, there was an entry for dactinomycin (COSMEGEN). Although an order for dactinomycin may have been appropriate since this was an oncology patient, the lack of a daptomycin entry signaled the possibility that dactinomycin had been given in error. Upon further investigation, the pharmacist found that, indeed, there was no order for dactinomycin. Most likely, the order entry pharmacist had, by mistake, selected a dactinomycin 500 mcg vial from the list of drugs on the computer screen instead of the intended daptomycin 500 mg.
Even though the correct drug had been dispensed, the wrong drug and strength (mcg instead of mg) had been entered into the patient's drug profile and on the subsequent printed product label. The nurse who administered the drug did not notice that the dispensed product had been mislabeled. Fortunately, no harm occurred in this case because the IV bag actually contained daptomycin, which was confirmed after interviewing the two pharmacists involved. However, if a pharmacist had not reviewed antibiotic usage that morning, the wrong drug could have been dispensed later and administered since all labels and reports, including any computer-generated medication administration record, would have listed dactinomycin, not daptomycin. Conceivably, similar errors could also happen with computerized prescriber order entry systems if the physician chooses the wrong look-alike drug name from a list presented on-screen. Dactinomycin labeling carries a number of warnings for both handlers and patients due to its high potential for toxicity, thus any error could cause serious harm.
Dactinomycin and daptomycin also sound alike when spoken, as mentioned in our October 30, 2003 newsletter. When a Cubist Pharmaceuticals (manufacturer of daptomycin) sales representative called a pharmacy director to discuss the company's antibiotic, telephone reception was poor and, at first, the pharmacy director thought he heard the sales representative say "dactinomycin." We've even heard some oncologists say "dactomycin" instead of "dactinomycin," which sounds even more similar to daptomycin. In addition to look-alike and sound-alike drug names and similar screen appearance of the 500 mg and 500 mcg vial strengths, each drug is a lyophilized powder that becomes a pale yellow solution upon reconstitution, and each is administered once daily.
Educating people about the risk of mix-ups between these products is necessary although not a long-term, high-leverage solution. In the past we've recommended storing chemotherapy drugs away from other medications, but that won't help if the wrong item is chosen from a computer screen and printed on a label. Instead, the manner in which drug information is communicated in computer systems needs modification.
For example, if your system allows, include both generic and brand names in listings, making the brand name primary and the generic name secondary for these two drugs. If generic names must be presented first, consider using tall man letters to highlight certain letter characters of each generic name to enhance recognition (DAPTOmycin and DACTINOmycin). Place warning labels on the bins where they are stored. Also build name-alert warnings in the pharmacy computer system and computerized prescriber order entry system. Be sure to verify the patient's diagnosis before dispensing these products. Incidentally, although these medications may not be used often, this is the third close call reported to us with these drugs. (One close call was recognized by an unexpectedly large dose difference.) To be certain that your organization would be able to detect this potentially harmful mix-up, test your system using the scenario described above. We've contacted Cubist Pharmaceuticals to alert them to the error potential so they can consider whether modification of the generic name for daptomycin may be warranted.
Draft label format guidelines for inpatient oral medications
In our February 2003 issue, we published a set of Draft Guidelines for Safe Electronic Communication of Medication Orders. Since then, a work team has been established with representation from ISMP, pharmacy practitioners, vendors (McKesson, Siemens, and Cerner), and the Joint Commission, to continue addressing this issue. Guided by human factors concepts and familiarity with computer system-related errors that have been reported to the USP-ISMP Medication Errors Reporting Program, the team's mission has been to facilitate the adoption of standard formats for programming pharmacy computer-generated outputs to reduce the risk of errors. The agreed-upon plan has been to start with safe formatting of medication container labels. The first set of guidelines covers labels for oral solid medications that are dispensed and administered in the inpatient setting. A draft of these guidelines is posted on our website at www.ismp.org/comments. We invite your review and encourage your comments (click on button at end of draft guideline document) on this important safety issue. Please respond by December 31, 2005. Final guidelines will be released during the first quarter of 2006. Future projects include label formats for inpatient oral liquids and parenterals, outpatient container labels, and electronic medication administration records.
Caution with slash marks. We've written before about errors with orders that contained slash marks. The table of error-prone abbreviations and symbols on our website (www.ismp.org/PDF/Error Prone.pdf) notes that slash marks are sometimes seen as the number 1. The order below for "5%D/0.2NSS," which includes the use of a slash mark, was nearly mistaken as dextrose 10%. It's doubtful that success will be achieved with prohibiting slash marks in all orders. So, we'll let this example above speak for itself and suggest that you share it with clinical staff to enhance awareness of the potential for such errors.
Oral liquid given IV. An order was written for "TUSSIONEX suspension" (hydrocodone and chlorpheniramine), but the oral route of administration was not expressed. Pharmacy dispensed the cough medicine in unit-dose oral syringes. However, the patient's nurse had recently joined the staff after working in a long-term care facility and was unfamiliar with oral syringes. She made the assumption that, since the liquid was in a syringe and the patient had IV access, the drug should be given IV. Unfortunately, a pharmacy label covered the words, "For oral use only," that the manufacturer had printed on the oral syringe. Noting that the drug was rather thick, she transferred it from the oral syringe into a regular syringe, diluted it with saline, and injected it. Afterwards, she commented to another nurse that the drug was quite sticky. Further queries led to recognition that the drug had been given by an incorrect route. The IV site was immediately removed, the patient was monitored, and fortunately, no harm occurred. Not every nurse is familiar with oral syringes and some, as in this case, may mistake a liquid medication in an oral syringe for a parenteral product. Awareness of oral syringes and how they protect against inadvertent IV administration should be included in all new staff orientation. Pharmacy labels should never cover important information, including "For oral use only" warnings. For added protection, auxiliary labels are available with this statement printed in large red characters, which can be affixed to the syringe plunger before dispensing an oral liquid in an oral syringe. Incomplete orders that do not express a route of administration also should be clarified before initial transcription of the orders.
Human factors issue. Healthcare providers and patients who use blood glucose meters from Abbott Diabetes Care (Alameda, CA) were notified in late October that, in the course of setting the time and date on the meters, at least 17 users inadvertently changed the units of measure from mg/dL (used in the US) to mmol/L (used in other countries). Similar changes in the measurement unit have also been reported after dropping the meter or replacing the battery. The affected Abbott glucose meters that are sold in the US are: FreeStyle, FreeStyle Flash, FreeStyle Tracker, Precision Xtra, Sof-Tact, Optium, MediSense, and private label brands ReliOn Ultima, Rite Aid, and Kroger meters. Precision Sof-Tact meters, which were inadvertently omitted from Abbott's press release, also are affected. Users in the US should make sure that their meter reading is displayed as mg/dL, as inaccurate readings could lead to unnecessary dietary changes or wrong doses of insulin, resulting in hyperglycemia. Although new meters that are currently being shipped to stores are locked with the correct unit of measure, be aware that old meters are still being sold and many are still in use at some hospitals. (We saw some older meters last week during a community pharmacy visit.) Abbott did not instruct current users to return the blood glucose meters for repair. If these devices are in use at your institution, you may want to attach a reminder to the device, or post a reminder at nursing units, about the correct setting. Also, alert nurses that blood glucose readings below 20 suggest that the device is expressing the value in mmol/L; the glucose meters do not read mg/dL values below 20. The company is still sending notices to registered users.
Your Reports at Work
Company warning: Omacor-Amicar mix-ups. Our medical director, Russell Jenkins, MD, recently received a communication from Reliant Pharmaceuticals warning about the potential for confusing OMACOR (omega-3-acid ethyl esters) prescriptions with AMICAR (aminocaproic acid). The letter can be viewed at: www.ismp.org/MS Aarticles/20051103_1.htm.Our November 3, 2005, newsletter reported such an incident, and at that time, we had asked the company to take action. Community pharmacies have received a similar letter. We appreciate the timeliness of Reliant's communication, however, we still feel that these two drug names and 1 g oral dosage strengths are far too similar. Thus, we stand by our recommendation to change the brand name, Omacor.
New video with free viewing. Produced by FDA in cooperation with ISMP, Caution on Accidentally Giving Nimodipine (NIMOTOP) Intravenously, is a new patient safety video available at: www.ismp.org/Pages/FDAVideos.htm. The video, which can be downloaded for free viewing, warns about inadvertently administering nimodipine IV, which has resulted in serious injuries and deaths. Other new medication safety videos from FDA include Stronger Risk Management Program for Accutane (isotretinoin); Recall of Certain Injectable Drugs by Central Admixture Pharmacy Services (CAPS); and Updated Cardiotoxicity Data for Herceptin (trastuzumab). These can be found using the link above, after clicking to view the nimodipine video.
Interested in color-coding? A transcript of the FDA Part 15 public hearing, Use of Color on Pharmaceutical Product Labels, Labeling, and Packaging, held last March at NIH in Bethesda, MD, is now available at: www.fda.gov/cder/meeting/part15_3_2005/Transcript.pdf. The views of various stakeholders, including ISMP, USP, AMA, ASHP, and the pharmaceutical industry are included in the transcript.
Does your facility provide inpatient or outpatient oncology services? If so, please help in overcoming the tragedy of misadministration of vincristine by taking a few minutes to complete our survey on safety of IV vincristine administration at: www.ismp.org/s/survey200512.asp. Also see lead article in this week's issue.
ISMP Medication Safety Alert! Acute Care (ISSN 1550-6312) ©2005 Institute for Safe Medication Practices (ISMP). Permission is granted to subscribers to reproduce material for internal newsletters or communications. Other reproduction is prohibited without written permission. Report medication errors to the USP-ISMP Medication Errors Reporting Program (MERP) at 1-800-FAIL-SAF(E). Unless noted, published errors were received through the MERP. ISMP guarantees confidentiality of information received and respects reporters' wishes as to the level of detail included in publications. Editors: Judy Smetzer, RN, BSN, Michael R. Cohen, RPh, MS, ScD, Russell Jenkins, MD. ISMP, 200 Lakeside Drive, Suite 200, Horsham, PA 19044. Tel. 215-947-7797; Fax 215-914-1492; E-MAIL: firstname.lastname@example.org. This is a peer-reviewed publication.
Updated August 11, 2008