Infusion Free-Flow Apparently Still a Risk
First reported in the late 1980s and early 1990s by ISMP and the ECRI Institute1-3, accidental, uncontrolled free-flow of IV solutions after removal of administration sets from infusion pumps was recognized as a serious problem that led to death and patient injury. Free-flow events occurred when pump tubing was removed from the device while still connected to the patient, before closing the set’s manual tubing clamp. This resulted in a large volume of solution infusing rapidly into the patient, along with drug or nutrient overdoses if the infusions contained medications or parenteral nutrition. In response to this risk, manufacturers introduced free-flow protected tubing as a fail-safe to prevent gravity flow, by automatically clamping the set immediately upon its removal from the pump. In 2001, The Joint Commission surveyors began evaluating healthcare organizations’ use of infusion pumps with free-flow protection, and in 2003, it implemented a National Patient Safety Goal (NPSG) to promote free-flow protection with all general-use and patient-controlled analgesia (PCA) IV infusion pumps in the organization.
This 2003 NPSG has, in essence, assured that manufacturers are no longer distributing non-protected devices. However, despite seemingly fail-safe interventions by device manufacturers, ISMP is aware of the potential for intravenous free-flow events if administration set tubing is misloaded in infusion pumps. Although subsequently addressed by the manufacturer, the problem was well documented in a 2006 article published in Anesthesiology, which described a free-flow event with the Alaris Medley Medication Safety System when the set was accidentally misloaded4. More recently, in April 2007, Smiths Medical also sent a Safety Alert letter to its customers to warn about the risk of a free-flow event with certain administration sets if the tubing is improperly attached to the pump. ECRI Institute also followed with a warning to its members about this risk. With two manufacturers reporting similar problems, we are concerned that this type of hazard may also be possible with other infusion pumps and administration sets.
The article published in Anesthesiology describes a free-flow event with a 58-year-old man undergoing surgery. The patient’s arterial pressure suddenly decreased during surgery for no apparent reason, requiring treatment with vasopressors. After the patient’s blood pressure was stabilized, staff realized that a nitroglycerin infusion had flowed into the patient via gravity and was empty. Upon close inspection, they noticed a gap at the top of the pump door and found that the upper flange part of the tubing had not been loaded properly. The door had closed sufficiently, but the misloaded tubing had disarmed the mechanism used to prevent free-flow. The pump database revealed that the pump had alarmed and displayed an error message during set-up. But the message reported a possible occlusion, thus the free-flow problem was not recognized and the alarm was overridden. In addition to design changes made in the pump module, the error message has since been changed in current software applications to better recognize the risk of free-flow.
While Smiths Medical has received no reports of free-flow events, the manufacturer has confirmed that, if certain administration sets (cassette reservoirs) with flow stop are incorrectly attached to its ambulatory pumps, CADD-1, CADD-Legacy, CADD-PCA, CADD-Plus, and CADD-Prizm, the flow-stop spring lever arm may not fully activate, potentially leading to unregulated gravity flow to the patient. Smiths Medical will be reintroducing the administration sets with integral anti-siphon valves to prevent further problems.
As mentioned, free-flow events due to misloading of administration sets may not be limited to these administration sets and infusion pumps. We highly recommend that you test your pumps in a simulated clinical setting to see if similar problems exist. You may also want to contact your infusion pump vendor to see if representatives are aware of free-flow risks when administration sets are misloaded in their pumps. The authors in the above-cited article4 emphasized the fact that they had performed an extensive failure mode and effects analysis (FMEA) before using the infusion pumps, but the FMEA did not predict the failure mode causing the free-flow event. The authors also noted that the FMEA team was aware of reports of incorrect loading of the tubing. Because of this, the team-designed training specifically focused on the correct loading of the administration set. Again, the training did not prevent the event. Please take the time to investigate how misloaded tubing could affect the performance of your infusion pumps. If you uncover any problems, please report them to ISMP (firstname.lastname@example.org). We will ensure that the reports are forwarded to FDA and the device manufacturer, and will publish confirmed problems in our newsletters.
- Cohen MR. Avoid the risk of electronic infusion control device free flow incidents. Hosp Pharm 1989:24:510.
- ECRI. Infusion pumps, general purpose. HealthDevices Alerts. FDA data accession No. 173580. December 22, 1989.
- Cohen MR, Davis NM. Free flow associated with electronic infusion devices: an underestimated danger. Hosp Pharm 1992; 27:387-90.
- Schroeder ME, Wolman RL, Wetternecek TB. Tubing misload allows free flow events with smart intravenous infusion pump. Anesthesiology 2006; 105:434-5.