Heparin errors continue despite prior, high-profile, fatal events
From the July 17, 2008 issue
To quote The Wall Street Journal, “Between overdoses and adulteration, the bad news about heparin keeps coming” (Francis T. Heparin overdoses hit babies in Texas hospital. July 9, 2008: http://blogs.wsj.com/health/2008/07/09/heparin-overdoses-hit-babies-in-texas-hospital/?mod=googlenews_wsj). Less than 2 weeks ago, as many as 17 infants in a Texas hospital neonatal intensive care unit (NICU) received heparin overdoses. Two infants, a set of premature twins, have since died, although their deaths have not been definitively linked to the errors. The other infants are in stable condition or have been discharged to home.
According to hospital press releases, the errors were set in motion on July 3 during the mixing process in the pharmacy. The erroneously prepared heparin products were first administered to infants in the NICU on the following day, July 4. On July 6, nurses discovered the error after investigating unexpected abnormalities during routine blood testing on several infants. Appropriate treatment was started to counter the adverse effects of heparin. Initially, the hospital’s chief medical officer reported that it was unclear how many infants received the incorrectly prepared heparin solution because there were syringes from a different drug batch also in the NICU. As of today, there are 14 confirmed cases and three possible cases. One pharmacist and two pharmacy technicians were on duty when the error occurred. According to the hospital, the two individuals involved in the error have taken voluntary personal leave.
Families with infants involved were informed about the error, but few details have been made public about the causes. Although the hospital’s press releases note that the error was unrelated to product labeling or packaging, it appears that the infants received as much as 100 times more heparin than intended. Based on our analysis of neonatal heparin dosing errors in other facilities, other potential causative factors for this type of error include:
Calculation error/unfamiliar concentration. The wrong concentration of heparin or wrong amount of diluent has been used in pharmacies to prepare solutions. These errors may arise from calculation errors and/or the presence of unfamiliar concentrations of heparin in the pharmacy. Following the recent heparin recall and subsequent shortage, many pharmacies have been forced to use heparin in concentrations different than they are accustomed to using. For example, a facility that normally uses heparin in 1,000 units/mL vials may be able to purchase heparin in only 10,000 units/mL vials; thus, the directions for preparing heparin products would need to be changed based on the new concentration. We discussed this before in our May 22, 2008 newsletter (www.ismp.org/Newsletters/acutecare/articles/20080522-1.asp).
Failed verification process. In some cases, erroneously mixed solutions have been dispensed from the pharmacy because product verification did not occur or the verification process failed to detect the error. These errors have been attributed to: 1) inadequate staffing, training, or documentation of compounding, 2) environmental factors, or 3) simple human error, which could have been detected if barcode technology had been employed.
Inconclusive industry standards. Studies on the use of heparin to maintain patency of venous and arterial lines in neonates yield conflicting data (www.ashp.org/s_ashp/docs/files/BP07/TPS_NaCl.pdf). Without definitive data about the safest and most efficacious practice for maintaining line patency, various protocols are currently being utilized for otherwise similar neonatal patients. Error-prone practices are inevitable when heparin in 0.5 units/mL, 1 unit/mL, 2 units/mL, or 10 units/mL concentrations, or saline alone (at least for peripheral IV lines with relatively larger bore catheters) are all seen as viable alternatives to maintain patency. Without a standard protocol, few manufacturers have been willing to offer premixed neonatal heparin products. More recently, preservative-free prefilled heparin flush syringes have been made available in 1 unit/mL, 2 units/mL, 10 units/mL, and 100 units/mL concentrations (e.g., www.bd.com/injection/products/posiflush/heparin.asp, www.excelsiormedical.com/excelsior/product_codes.asp, www.hospira.com/Files/040393Prefilled_Flush.pdf), which should help reduce the risk of pharmacy or nursing compounding errors when syringes are needed. (These prefilled syringes may be in short supply until the heparin shortage ends.) To reduce the risk of confusion between various concentrations of heparin in similar-looking vials, some hospitals use a prefilled syringe of preservative-free heparin 100 units/mL (500 units/5 mL) to prepare flushes. Heparin is also administered to neonates via continuous infusion through an umbilical vein or artery, but no premixed solutions with low concentrations of heparin are commercially available for this use.
To give credit where credit is due, the hospital where the most recent error occurred had implemented an important safety practice to prevent heparin errors before the event: pharmacy routinely dispensed heparin flush solutions for neonates instead of having nurses prepare the solutions. Although the risk of errors is lessened, this event clearly demonstrates that errors are still possible when pharmacy prepares heparin flush products. So we need to do more to protect our tiniest patients.
Barcode technology employed during the mixing process is one solution. A hospital pharmacist recently told us about a different solution: a refractometer is used to measure the specific gravity of neonatal heparin solutions mixed in batches or for individual patients. Because the heparin concentration in neonatal solutions is so low, the refractometer reading is expected to be the same as the reading for the diluent alone. If not, an error in the heparin concentration is suspected and investigated. Outsourcing the preparation of neonatal heparin flush products is another option. These commercial manufacturers typically run assays on batched solutions, and although the heparin concentration is, again, too low to measure, one can be assured that an abnormally high concentration of heparin hasn’t been prepared. Direct medication validation technology such as ValiMed (www.valimed.com/index.htm), which uses enhanced photoemission spectroscopy to check accuracy, is also available (Kaakeh Y, et al. Enhanced photoemission spectroscopy for verification of high risk IV medications. Am J Health-Syst Pharm 2008;65:49-54).
It’s frustrating and tragic that heparin errors continue to happen and harm neonates. At this point, every hospital in the nation should be conducting a failure mode and effects analysis regarding the use of heparin in neonates, both in flush solutions and when added to infusate. We also need to think ahead and consider whether our solutions today will still be effective in tomorrow’s healthcare systems.
This week in Rockville, MD, a number of our colleagues met at an invitational summit on IV drug safety hosted by the American Society of Health-System Pharmacists, with co-conveners ISMP, The Joint Commission, United States Pharmacopeia, The National Patient Safety Foundation, and The Infusion Nurses Society (www.ashp.org/s_ashp/doc1c.asp?CID=489&DID=7933#expert). We expect a set of far-reaching recommendations associated with the use of IV products that will positively affect patient safety. If you have discovered an innovative way to prevent medication errors—particularly heparin errors—we would love to hear from you!
If you need assistance...
Interdisciplinary consultant staff from ISMP are available to assist hospitals in performing a proactive risk assessment of their medication system, with special focus, if requested, on IV therapy—from prescription to administration. For information, call Michelle Mandrack, RN, Director of Consulting Services, at 215-947-7797, or visit: www.ismp.org/consult/default.asp.