|
EMS House of DeFrance http://www.emshouse.com Studies Trials Abstracts Courtesy the EMS House of DeFrance http://www.defrance.org Source: Prehospital Emergency Care ABSTRACT
Objective. Establishing traditional intravenous (IV) access in adult trauma and medical patients can be difficult. We evaluated provider performance for obtaining intraosseous access with two FDA- approved intraosseous devices (F.A.S.T. 1(TM) and EZ-IO) in two sequential field trials. Methods. One hundred twenty-four providers consented to participate in the first field trial evaluating the use of the F.A.S.T. 1(TM) system. Three hundred eighty-nine providers consented to participate in the second field trial, evaluating the use of the EZ-IO. Following each insertion attempt, a telephone data collection process with a member of the research team was completed. Insertion success rate and measures of provider comfort and satisfaction with each device were collected and analyzed. Results. One hundred seventy-eight insertions (89 F.A.S.T. 1(TM) ; 89 EZ-IO) were completed between February 2000 and December 2005. Sixty-four of the 89 insertions of the F.A.S.T. 1(TM) were successful, and 78 of the 89 insertions of the EZ-IO were successful (72% vs. 87%; ÷^sup 2^ = 6.8; p = 0.009). Providers using the F.A.S.T.1(TM) attempted more IV insertions prior to using the IO device than the providers using the EZ-IO (2.6 vs. 2.0, p = 0.005). There were no differences in provider comfort or provider assessed device performance between the two devices (p = 0.52; p = 0.13, respectively). Conclusion. In our comparison of two field trials of prehospital provider use of the F.A.S.T.1(TM) and EZ-IO systems, more successful insertions with the EZ-IO were achieved than with the F.A.S.T.1(TM) device. Limitations of our comparison include nonrandomization, the sequential field trial design, the potential for a learning effect, and self-reporting of data points by providers. A prospective, randomized evaluation of these devices is warranted to draw definitive conclusions about provider insertion success rate with these devices. Key words: intraosseous; IO; IV access; fluid infusion; intravenous failure; emergency medical services; emergency medicine.
PREHOSPITAL EMERGENCY CARE 2007;11:164-171
INTRODUCTION
Prehospital personnel routinely encounter patients who require emergent initiation of intravenous (IV) lines for the administration of fluids, either to replace lost volume or for medication administration. The IV failure rate reported in the literature is approximately 10%.1-6 A recent audit of 300 critically ill or injured patients treated by services under our medical direction showed a 9% IV failure rate. That means, despite one to six IV attempts, 27 patients who were in need of an IV line for volume replacement or medication administration did not receive one.
When venous access is not available to providers, medication administration via the endotracheal tube (ETT) has been used. However, this is a limited option, because not all medications can be administered through the ETT, only small volumes of fluid can be delivered, and there is concern about its efficacy.7 The pediatric population, especially those under age five, is especially susceptible to IV access difficulties because of small vein size. For years, paramedics have used the alternative of intraosseous (IO) access for this patient population. Although there is literature reporting the use of IO access in adults, it is still relatively rare.8-10
Because of the poor alternative treatments for failed IV attempts and the lack of information on adult intraosseous devices, we decided to evaluate provider performance on the use of an PDA- approved, sternal intraosseous infusion device (F.A.S.T. 1(TM) IO Infusion System, Pyng Medical, Richmond, British Columbia, Canada). This device consists of an array of needles that surround a central needle through which a plastic catheter passes (Figure 1). When the device is placed on the manubrium (Figure 2), the array of needles marks the depth of the periosteum. When approximately 60 PSI is manually applied to the device, the central needle penetrates the bone to a depth of 6 mm. The device is then withdrawn and the catheter stays in place.
Following that field trial, a new tibial intraosseous infusion device was approved by the FDA for use in adult patients. This device, the EZ-IO (Vidacare, San Antonio, TX), was subsequently field trialed by providers in our system. This device consists of a 25-mm, 15-gauge needle set divided into two parts: the catheter and a central stylet powered by a 12-volt battery operated electric drill (Figure 3). With the drill operating, the needle set is advanced through the periosteum into the medullary space one finger width distal and medial to the patient's tibial tuberosity (Figure 4). The stylet is then withdrawn and the catheter stays in place. Both devices are aspirated and flushed in the usual fashion to check for proper placement and then infused under 300 mmHg of pressure.
Data for insertion success rates (the primary end point), average number of IV attempts prior to IO attempt, estimated time to fluid infusion, complications, provider comfort with the device, and provider evaluation of device effectiveness were collected during both field trials and analyzed. The success rates for first time insertions, as well as success rates for providers who completed multiple insertions, were also analyzed for a possible learning effect for the studied devices.
METHODS
Both field trials were approved separately by the HealthPartners Research Foundation Institutional Review Board. Both field trials were designed, reviewed, and approved by our IRB as an evaluation of the providers as the study subjects, which did not require informed consent from the patients receiving an intraosseous device. The first field trial consisted of the training, use, and evaluation of the F.A.S.T. 1(TM) IO Infusion System, and the second field trial consisted of the training, use, and evaluation of the EZ-IO. Emergency medical services from the East Metropolitan area of the Twin Cities (Minneapolis and St. Paul, MN) and three counties in western Wisconsin were invited to participate in both phases of the study. Participation by the providers was completely voluntary. The participants were from both full-time and volunteer agencies and provided 9-1-1 response to urban, suburban, and rural communities, as well as interfacility critical care transfers.
Field Trial of the F.A.S.T1(TM)
All EMT-Ps, RNs, and EMT-Is employed by the services expressing interest in the research project attended an initial training session. This 2.5-hour session consisted of a PowerPoint presentation, followed by a written exercise completed by each provider, a discussion about the consent process, and a hands-on insertion simulation with the device. EMS instructors from our institution conducted all of the training sessions. All of those in attendance at the training sessions consented to participate in the study-a total of 124 providers.
Inclusion and exclusion criteria for use of the F.A.S.T. 1(TM) are listed in Table 1. If the patient met the inclusion criteria, the guideline for use of the F.A.S.T. 1(TM) was after one failed IV attempt, or if intravenous access was anticipated by the provider to be difficult or impossible. In our system, an on-call medical direction coordinator is available 24/7 for medical direction needs, and these two individuals were used to complete the study data collection. Following each insertion attempt, the provider paged the on-call study coordinator, who called the provider back immediately and completed the data collection form over the telephone.
Field Trial of the EZ-IO
Immediately following the conclusion of the F.A.S.T. 1(TM) field trial, the field trial of the EZ-IO began. As with the F.A.S.T. 1(TM), all EMT-Ps, RNs, and EMT-Is employed by the services expressing interest in participating in the research trials attended an initial 1.5-hour training session, which consisted of a PowerPoint presentation about the use of the device, a written evaluation following the PowerPoint presentation, a discussion about the consent process, and a hands-on insertion simulation using the device. The EMS instructors from our institutions completed all of the training sessions. On successful completion of the training session, the devices were deployed for use in the field. Inclusion and exclusion criteria for use of the EZ-IO are listed in Table 1. Similar inclusion and exclusion criteria for entry into the study were used for both devices; however, the PDA-approved use of the EZ- IO includes patients who fall outside of these criteria. Providers were instructed to use the EZ-IO according to the PDA-approved indications for use, with an understanding by the investigators that only those cases that met the study inclusion criteria would be analyzed. By design and to attempt a more fair comparison, the same guideline for use of the F.A.S.T.1(TM) applied to this field trial (one failed IV attempt, or intravenous access that was anticipated by the provider to be difficult or impossible).
Following every insertion attempt, the provider contacted the radio control center under our medical direction and completed the same data collection form (Table 2) as was completed for the F.A.S.T. 1(TM) field trial. The on-call study coordinator was available 24/7 to address any problems experienced by providers.
At the completi\on of both field trials, the data were double entered into a database and cleaned for data entry errors. Successful placement of the device was defined as penetration of the bone by the needle followed by successful fluid infusion. Estimated time to infusion was evaluated on the basis of the provider's estimate from the time it took to pull the device from the package to the time FV fluid was infused. A first attempt was defined as the first use of a device by a provider. Experienced providers were those who made two or more attempts with the EZ-IO or the F.A.S.T.l(TM). Those who made one attempt with the EZ-IO and one attempt with the F.A.S.T.1(TM) were not considered experienced providers due the large differences in insertion technique and location.
A ÷^sup 2^ test was used to compare overall insertion success rates between the two devices. A two-sample proportion test was used to determine if a difference existed between first time insertion success rates and success rates for multiple insertion attempts. A comparison of first time insertion success rates between the F.A.S.T.1(TM) and the EZ-IO , as well as a comparison between the success rates of providers with multiple insertions with the F.A.S.T.1(TM) and EZ-IO (experienced providers), was completed by using a two-sample proportion test. Time to fluid infusion, provider comfort, and device effectiveness were analyzed by using the Cochran- Mantel-Haenszel Test of General Association. A two-sample t-test was used to compare the number of IV insertion attempts between the two devices.
RESULTS
Of the 124 providers consenting to participate in the F.A.S.T. 1(TM) field trial, 55 completed at least one insertion (44%). Of the 385 providers consenting to participate in the EZ-IO field trial, 63 (18%) had at least one insertion. One hundred seventy-eight insertions (89 F.A.S.T. 1(TM), 89 EZ-IO) meeting the inclusion criteria in Table 1 were completed by providers between February 2000 and December 2005. Patient demographic information is available in Table 3. Table 4 shows the overall insertion success rate, first attempt success rate, and providers with multiple insertion (2 or greater insertions) success rate for both devices. The overall success rate was higher for the EZ-IO than the F.A.S.T. 1(TM) (87% vs. 72%; ÷^sup 2^ = 6.8; p = 0.009). There was no difference between the two devices with respect to first attempt success rate (p = 0.112) or providers with multiple attempts (p = 0.159). There was also no difference in success rate between first attempt providers and multiple attempt providers for either the EZ-IO (p = 0.510) or the F.A.S.T. 1(TM) (p = 0.493).
As seen in Table 5, the reported time to fluid infusion with the EZ-IO was faster than the reported time to infusion for the F.A.S.T. 1(TM) phase (p = 0.02). No differences in provider comfort or device effectiveness were detected between the two devices (p = 0.52 and 0.13, respectively). Providers using the F.A.S.T. 1(TM) attempted more IV insertions prior to using the IO device than the providers using the EZ-IO (2.62 vs. 2.08; p = 0.005).
DISCUSSION
Our success rate with the F.A.S.T. 1(TM) was quite different from success rates previously reported in the literature.10 The reason for this is unclear, but our experience was consistent throughout the duration of the field trial. The insertion failures common to both devices were excessive tissue over the insertion site, successful penetration of the bone but an inability to infuse through the device, and fluid extravasation around the insertion site (Table 6). Development of a longer needle for the bariatric population is needed to prevent insertion failures due to excessive tissue over insertion sites for both devices. Inability to infuse following successful penetration of the bone has several possible explanations. One explanation may be the process of aspirating an IO needle prior to infusion. This attempt to confirm placement also has the potential to plug the needle or catheter with marrow or bone. In clinical practice, it has been our impression that if bone marrow aspiration fails and the device subsequently infuses without extravasation, the device is still used. This appears to be the case especially in critical patients.
As seen in Table 6, the most common reasons for failure of the F.A.S.T. 1(TM) were problems with the catheter dislodging prior to fluid infusion, or the catheter stretching and breaking after initial penetration of the bone. In these cases, fluid flow through the device was impossible. These insertions were considered failures according to our definition of insertion success as penetration of the bone with successful fluid flow through the device. It is our sense that the placement of this device requires the provider to use a precise insertion technique throughout the entire procedure. The device must be placed directly perpendicular to the manubrium. The wrist and the elbow need to be in linear alignment and the device must not be torqued. If any of these aspects of the technique are not followed, there is an increased risk of failure. The EZ-IO appears to be superior to the F.A.S.T. 1(TM) in insertion success rates, which may be attributed to the more precise insertion technique required by the F.A.S.T. 1(TM).
With the EZ-IO , providers reported problems related to drill power or battery failure. Several times providers reported a "binding" phenomenon, where the drill would slow and eventually stop during the insertion attempt. This problem was widely reported but often did not result in a failed attempt, as only one insertion failure was attributed to this problem. We believe there are three possible explanations for this occurrence. First, if the provider pushed the drill too hard against the bone while attempting an insertion, the needle would not have enough rotation speed to penetrate the bone. This situation is analogous to pushing too hard with a power drill when attempting to place a screw into a piece of wood. This issue was addressed with additional training to providers. Another explanation is the use of the wrong brand of battery. The manufacturer informed us that AA Duracell(TM) batteries (Proctor & Gamble, Bethel, CT) are approximately 1 mm longer than other brands and that the extra length maintains a more consistent contact within the battery pack. And finally, even with the correct batteries in place, it was the feeling of the majority of providers that the drill was underpowered. A new lithium ion battery pack is now available for use and provides 10% more torque than the alkaline batteries used in the old battery packs.
Removal of either of these devices is quite simple. The F.A.S.T 1(TM) device is supplied with an extraction tool which is required for removal. The tool has threads on the distal end and is inserted into the metal end of the catheter, which contains a matching set of threads. The tool is then threaded onto the catheter and the catheter is pulled out. The EZ-IO device has a Luer-Lock end to which a syringe is attached. The device is then removed with a clockwise rotation and pulling morion. There was one removal failure with each of these devices. In one case, the metal catheter end of the F.A.S.T 1(TM) broke off and the catheter had to be removed with a locking forceps. The EZ-IO removal failure was the result of the plastic hub breaking off. The remainder of the device was also successfully removed with a forceps.
The primary limitation of our study is the fact that the initial field trials was not designed with the intent to compare the performance of the F.A.S.T. 1(TM) with the EZ-IO because the EZ-IO was not commercially available at the time of the first field trial. We do believe that subsequent research evaluating the use of these two devices in a more controlled, prospective fashion is needed. Other limitations of this study include the possible order effect of the study design, the self-reporting of data, specifically in regard to the time to fluid infusion data point, and the duration of the study period (6 years with potential for continuing education to influence success or changes in the provider attitude toward use of one or both devices).
The EZ-IO has FDA approval for broader usage on patients, some of whom fell outside of our study inclusion criteria. Providers were allowed to use the device according to the PDA-approved indications for use, with the understanding by the investigators that these cases would not be included in the field trial comparison. This resulted in the use of the EZ-IO several times on patients outside of the study protocol (Table 1). In an effort to maintain a comparable study sample, the same inclusion criteria for age and GCS developed for the F.A.S.T. 1(TM) trial were used for the EZ-IO trial, and only those insertions that met these criteria were used for this analysis. These cases were evaluated at the conclusion of our study and present some interesting results we believe are worthy of discussion. The EZIO was used nine times on patients under the age of 18 years old (range 9-18). It was successful 100% of the time (9/9). Twenty-three insertion attempts were made on patients with a GCS > 6, 20 of which resulted in a successful insertion (87%). If these 32 patients had been included from the data analysis, the success rate of EZIO insertions in this trial would have increased from 87% to 88% (107/121).
The authors believe that the ability to freely use the EZ-IO on all patients, regardless of GCS, makes this device unique and much more useful. The administration of 50 mg of lidocaine following the initial flush for pain control was used on those patients who had a GCS greater than 6. Administration of additional pain control medication may be needed, because we discovered that 50 mg lidocaine did not effectively control pain for some patients. In the past, the use of IO w\as felt by some to be expensive and most often associated with negative patient outcome (i.e., death). Although a formal cost-benefit analysis was beyond the scope of this field trial, the expanded capability of using this device on a broad based patient population may make the technology more acceptable to some providers in that regard. There is no reason that the F.A.S.T. 1(TM) could not be used on such patients, but because of the insertion location and technique, we do not believe that its use on this classification of patients is acceptable to either providers or patients. The successful use of the EZ-IO in pediatric patients leads the authors to believe this technology should be expanded to include that age range. The company has obtained FDA approval for a pediatric needle, which is currently being evaluated by providers in our EMS systems.
We believe an important area for future research is a comparison between the use of central lines and IO in the emergency department, specifically focusing on infusion rates, central circulation times, and complication rates.11-13
CONCLUSION
In our comparison of two field trials of prehospital providers use of the F.A.S.T. 1(TM) and EZ-IO systems, more successful insertions with the EZ-IO were achieved than with the F.A.S.T. 1(TM) device. Limitations of our comparison include nonrandomization, the sequential field trial design, the potential for a learning effect, and self-reporting of data points by providers. A future prospective, randomized evaluation of these devices is warranted to draw definitive conclusions about provider insertion success rate.
The authors thank the EMS providers from the following services for their participation in the research study: Apple Valley- LakevilleFarmington Ambulance, Burnsville Fire, Cottage Grove Police and Fire, Ellsworth Area Ambulance, Hastings Fire, HealthEast Transportation Service, Lakeview Ambulance, LifeLink III, Mahtomedi Fire, Maplewood Police and Fire, Oakdale Fire, Red Wing Fire, St. Croix EMS, St. Paul Fire and Safety Services, Unity Ambulance, White Bear Lake Fire, Woodbury Police and Ambulance, and Lakes Region EMS. They also thank the operators in the East Metro MRCC for their help in the data collection process and Sara Krohn and Thorn Flottemesch for their assistance with the statistical analysis.
References
1. Slovis CM, Herr EW, Londorf D, et al. Success rates for initiation of intravenous therapy en route by prehospital care providers. Am J Emerg Med. 1990;8(4):305-7.
2. O'Gorman M, Trabulsy P, Pilcher DB. Zero-time prehospital IV. J Trauma. 1989;29(l):84-6.
3. Cwinn AA, Pons PT, Moore EE, et al. Prehospital advanced trauma life support for critical blunt trauma victims. Ann Emerg Med. 1987;16(4):399-403.
4. Jones SE, Nesper TP, Alcouloumre E. Prehospital intravenous line placement: a prospective study. Ann Emerg Med. 1989;18(3):244- 6.
5. Spaite DW, Valenzuela TD, Criss EA, et al. A prospective in- field comparison of intravenous line placement by urban and nonurban emergency medical services personnel. Ann Emerg Med. 1994;24(2):209- 14.
6. Pons PT, Moore EE, Cusick JM, et al. Prehospital venous access in an urban paramedic system-a prospective on-scene analysis. J Trauma. 1988;28(10):1460-3.
7. American Heart Association. Part 7.2. Management of cardiac arrest. Circulation. 2005;112:24:IV-58-66.
8. Kruse JA, Vyskocil JJ, Haupt MT. Intraosseous infusions: a flexible option for the adult or child with delayed, difficult, or impossible conventional vascular access. Crit Care Med. 1994;22(5):728-9.
9. LaRocco BG, Wang HE. Intraosseous infusion. Prehosp Emerg Care. 2003;7(2):280-5.
10. Macnab A, Christenson J, Findlay J, et al. A new system for sternal intraosseous infusion in adults. Prehosp Emerg Care. 2000;4(2):173-7.
11. Davidoff J, Fowler R, Gordon D, et al. Clinical evaluation of a novel intraosseous device for adults. JEMS. 2005;30(10):s20-23.
12. Miller, L, Kuhn, }, Von Hoff, D. Does IO equal IV? Unpublished report. Available at: http://www.vidacare.com/reports/ DoesIOequaliv.pdf. Accessed April 11, 2006.
13. Neufeld JD, Marx JA, Moore EE, Light AI. Comparison of intraosseous, central, and peripheral routes of crystalloid infusion for resuscitation of hemorrhagic shock in a swine model. J Trauma. 1993;34(3):422-8.
Ralph J. Frascone, MD, FACEP, Joe P. Jensen, RN, NREMT-P, Kory Kaye, MD, FACEP, Joshua G. Salzman, MA, EMT-B
Received June 20, 2006, from Regions Hospital Emergency Medical Services, Regions Hospital, St. Paul, MN (RJF, JPJ, KK, JGS). Revision received October 3,2006; accepted for publication November 8, 2006.
Abstract presented at the 2006 Annual Meeting of the National Association of Emergency Medical Services Physicians, Tucson, AZ.
All EZ-IO drivers and three needles per driver set were donated by Vidacare, Inc. for use in this trial. The F. A.S.T. 1IO sets were purchased from Pyng, Inc. at a 50% discounted rate for use in this trial. No other consideration was received from either company.
Address correspondence and reprint requests to: Josh Salzman, Regions Hospital EMS, 640 Jackson St. MS: 13801B, St. Paul, MN 55101. E-mail: doi: 10.1080/10903120701205851
Copyright Taylor & Francis Ltd. Apr-Jun 2007
(c) 2007 Prehospital Emergency Care. Provided by ProQuest Information and Learning. All rights Reserved.
By utilizing the content on this page, you agree to the legal terms The
contents, unless otherwise specified, are copyrighted by ©
EMS House of DeFrance. |