Trending Topics

Prove it: IV access increases out-of-hospital cardiac arrest survival better than IO access

A recent study shows an increased survival rate for out-of-hospital cardiac arrest patients who are administered medication through IV access rather than IO access


A recent study shows an increased survival rate for out-of-hospital cardiac arrest patients who are administered medication through IV access rather than IO access.

Photo/Wikimedia Commons

By Kenneth W. Navarro

Case Review: Medic 14 and Engine 27 respond to a report of a patient not breathing in a bank lobby. Before arriving on the scene, the crews learn that bystanders are performing CPR.

The team arrives to find a male lying supine on the floor with a security officer and a nurse performing CPR. The AED on the scene delivered two shocks and the third analysis indicated no shock was necessary. The firefighters take over CPR while Medic Hernandez attaches the ECG pads. The rhythm on the monitor displays asystole.

One of the firefighters prepares a saline bag and administration set. Medic Sanders exposes the patient’s lower left leg and prepares to insert an intraosseous (IO) needle and catheter. With little difficulty, Sanders establishes the IO line, verifies patency and secures the tubing. She administers 1 mg epinephrine and flushes the IO line. After ten minutes and a second dose of epinephrine with no change in the ECG rhythm, the crew prepares to transport.

The patient remains asystolic as the crew arrives in the emergency department (ED). The ED team continues the resuscitation attempt for about 15 minutes before stopping.

Later at the station, the fire and EMS crew are discussing the case. Medic Sanders comments that, even though it did not make a difference in this case, she prefers IO access as the primary vascular access site for cardiac arrest because it is easy and it allows her to deliver medications sooner than the IV route, which she believes generally improves outcomes.

IV vs. IO study review

Researchers with the Resuscitation Outcomes Consortium examined the association between vascular access routes and outcomes following out-of-hospital, non-traumatic cardiac arrest in the adult patient [1]. Analysists separated patients who qualified for this investigation into two groups: those with intravenous (IV) access and those with IO access. Patients were not eligible for enrollment if the IV or IO attempts were unsuccessful or if both access routes existed.

The primary outcome variable in this investigation was survival to hospital discharge with favorable neurologic status. The authors used the definition of favorable neurologic status (Rankin scale score of 0-3) recommended by the International Liaison Committee on Resuscitation [2]. As secondary outcome variables, the researchers measured rates of return of spontaneous circulation (ROSC) and survival to hospital discharge. Secondary outcome variables are often included in research studies and, although they are not the primary focus, they can provide evidence supporting the primary outcome variable as well as providing some rationale for future research projects [3].

The data analysis involved three different procedures. First, the team performed a series of regression analyses to control mathematically all measured variables that could have influenced the outcome of the study. This strategy permits isolation of the effects of the two vascular access routes on the primary outcome variable. The second analysis procedure involved a statistical comparison technique called propensity matching. This technique uses the available data to match individuals from each vascular access group who have similar characteristics. Propensity matching allows a fair comparison by attempting to ensure the two groups are not significantly different from each other, except that one group had IV access while the other had IO access.

Finally, the researchers accounted for missing data through a statistical process called imputation, which can accurately estimate missing data values. In the final analysis, the researchers performed 50 imputations for each missing value and used the average value for the analysis. Collectively, these different statistical procedures helped increase the precision of the inference drawn from the outcome data.

What the IV vs. IO study results mean for you

During the study period, 115 EMS agencies responded to 17,445 non-traumatic, out-of-hospital cardiac arrest cases. After excluding 4,290 cases that did not meet the inclusion criteria, the researchers performed the analysis on the remaining 13,155 cases. The overwhelming majority of the study participants (95 percent) received IV access. Survival with favorable neurologic outcome (primary outcome variable) was much lower for the IO group (1.5 percent) than for the IV group (7.6 percent). Similarly, the IO group had lower ROSC rates (23.9 percent vs. 38.3 percent, respectively) and lower survival to hospital discharge rate (3.8 percent vs. 10.3 percent, respectively) compared to the IV group.

Results of the multivariable regression models indicate that when prehospital care providers utilize IO instead of IV as the primary vascular access site, adult patients who suffer non-traumatic out-of-hospital cardiac arrest are 40 percent less likely to achieve ROSC, 56 percent less likely to survive to hospital discharge, and 76 percent less likely to have a favorable neurologic status at discharge.

Every statistical analysis procedure utilized in this investigation returned the same results. EMS use of IO access reduced the probability of achieving ROSC, and having favorable neurologic status at hospital discharge.

Before you throw away all your IO needles, you should note the disappointing results associated with IO access could reflect a selection bias, i.e., medics did not use the same selection process for establishing the two groups. For example, it is possible that medics chose IO insertion when it seemed the patient was less likely to respond to the resuscitation attempt. In support of this possibility, demographic data demonstrated the IO group had fewer bystander-witnessed cardiac arrests and was more likely to present in a non-shockable rhythm. Both variables are associated with worse outcomes.

However, propensity score matching attempts to control selection bias by allowing researchers to compare only individuals from each group with similar characteristics. Those characteristics can include age, sex, initial rhythm presentation, witness status and whether a bystander performed CPR, just to name a few. Propensity score matching creates groups with essentially the same characteristics, making comparisons more fair. In this case, the research team matched 558 patients in the IV group to 558 patients with similar characteristics in the IO group. After matching, the results of the analysis were almost identical to the earlier regression analysis results. Patients who received an IO were 77 percent less likely to survive with favorable neurological outcome when compared to patients with an IV.

Finally, the analysis of the 50 imputed datasets confirmed the results of both earlier analyses. Patients in the IO group were 78 percent less likely to survive with favorable neurological outcome compared to the IV group.

Although the present study is the largest investigation into the association between vascular access routes and outcomes in this category of medical emergency, it is not the only study to reach this conclusion. A recent retrospective study of 1,800 patients suffering non-traumatic, out-of-hospital cardiac arrest in King County, Washington found those treated with an IO instead of an IV were less likely to achieve ROSC or to survive long enough to be admitted to the hospital, although survival to hospital discharge rates were not different [4]. Similarly, a retrospective review of more than 1,300 cases of out-of-hospital cardiac arrest conducted in New York State found patients were less likely to achieve ROSC if the medics administered the first dose of epinephrine through an IO site compared to an IV site, although the difference was not statistically significant [5].

With a number of studies coming to the same conclusion, it is reasonable to wonder why IO access might impair survival outcomes compared to IV access. Part of the reason could be time-to-treatment differences between the groups. Many places consider IO placement a backup venous access point when IV attempts are not possible or have failed. This approach would result in later rather than earlier vascular access, which could delay the administration of resuscitation medications.

One must also consider whether anatomical factors are responsible for reduced survival metrics associated with IO access. An evidence-based review using animal models of cardiac arrest found increased and faster-appearing serum concentration of epinephrine with IV administration compared to IO administration [6]. To a large degree, absorption of intraosseous epinephrine is determined by blood flow through the medullary cavity [7]. For patients in hemorrhagic shock, blood flow through the bone marrow decreases by about 30 percent and this decrease may be even greater for patients in cardiac arrest [7]. In addition, the pressure gradient between the inferior vena cava and the right atrium is virtually nonexistent during CPR, which severely limits blood return to the heart from below the diaphragm [8]. Thus, there may be significant delays in resuscitation medications reaching the heart when introduced into the circulatory system from IO access sites in the lower extremities [9].

Limitations of the present IV vs. IO study

There are a number of limitations in this trial that could affect the generalizability of the results. First, the researchers designed the present study to examine data collected for a different study altogether. This prevented the research team from controlling any of the variables that could affect the outcome. For example, the dataset did not identify specific IO devices or information about the anatomic location of the IO attempt, which can produce significant variation in fluid flow rates [10] and time to reach peak arterial blood concentration of epinephrine [11]. The dataset also did not contain information about whether the medics provided a fluid flush after epinephrine administration or used a pressure bag. Both of these interventions can influence drug distribution [11, 12].

As previously mentioned, the paramedics may have inadvertently introduced selection bias through their decision of who would receive an IV and who would receive an IO. The researchers attempted to control any potential bias with propensity score analysis, but it is possible that unknown or unmeasured variables may have confounded the results.

Finally, there is some evidence to suggest that survival benefits produced by epinephrine may be related to administration timing, i.e., earlier administration of epinephrine is associated with better increased survival metrics compared to later administration [13], although giving the drug too early may also be detrimental [14]. The dataset used for this analysis did not include information about how quickly the medics established vascular access or how quickly they administered resuscitation drugs. Thus, it is impossible to know what effect epinephrine administration timing had on these results.

Summary findings

The results of this investigation suggest that IV access rather than IO access may provide survival advantages for adults who suffered a non-traumatic, out-of-hospital cardiac arrest. Although this study was recently published, the original dataset used for this analysis was collected about a decade ago when some IO devices were relatively new. While only 5 percent of the patients in this study received IO access, recent studies have reported IO prevalence rates of 15 percent and 40 percent in patients suffering from out-of-hospital cardiac arrest [4,5] despite the lack of strong evidence demonstrating increased survival.

The author has no financial interest, arrangement, or direct affiliation with any corporation that has a direct interest in the subject matter of this presentation, including manufacturer(s) of any products or provider(s) of services mentioned.


  1. Kawano, T., Grunau, B., Scheuermeyer, F. X., Gibo, K., Fordyce, C. B., Lin, S., Stenstrom, R., Schlamp, R., Jenneson, S., & Christenson, J. (2018). Intraosseous vascular access is associated with lower survival and neurologic recovery among patients with out-of-hospital cardiac arrest. Annals of Emergency Medicine, 71(5), 588-596. doi:10.1016/j.annemergmed.2017.11.015
  2. Perkins, G. D., Jacobs, I. G., Nadkarni, V. M., Berg, R. A., Bhanji, F., Biarent, D., Bossaert, L. L., Brett, S. J., Chamberlain, D., de Caen, A. R., Deakin, C. D., Finn, J. C., Gräsner, J. T., Hazinski, M. F., Iwami, T., Koster, R. W., Lim, S. H., Ma, M. H., McNally, B. F., Morley, P. T., Morrison, L. J., Monsieurs, K. G., Montgomery, W., Nichol, G., Okada, K., Ong, M. E., Travers, A. H., & Nolan, J. P. (2015). Cardiac arrest and cardiopulmonary resuscitation outcome reports: Update of the Utstein Resuscitation Registry templates for out-of-hospital cardiac arrest: A statement for healthcare professionals from a Task Force of the International Liaison Committee on Resuscitation (American Heart Association, European Resuscitation Council, Australian and New Zealand Council on Resuscitation, Heart and Stroke Foundation of Canada, InterAmerican Heart Foundation, Resuscitation Council of Southern Africa, Resuscitation Council of Asia); and the American Heart Association Emergency Cardiovascular Care Committee and the Council on Cardiopulmonary, Critical Care, Perioperative and Resuscitation. Resuscitation, 96, 328-340. doi:10.1016/j.resuscitation.2014.11.002
  3. Vetter, T. R., & Mascha, E. J. (2017). Defining the primary outcomes and justifying secondary outcomes of a study: Usually, the fewer, the better. Anesthesia and Analgesia, 125(2), 678-681. doi:10.1213/ANE.0000000000002224
  4. Feinstein, B. A., Stubbs, B. A., Rea, T., & Kudenchuk, P. J. (2017). Intraosseous compared to intravenous drug resuscitation in out-of-hospital cardiac arrest. Resuscitation, 117, 91-96. doi:10.1016/j.resuscitation.2017.06.014
  5. Clemency, B., Tanaka, K., May, P., Innes, J., Zagroba, S., Blaszak, J., Hostler, D., Cooney, D., McGee, K., & Lindstrom, H. (2017). Intravenous vs. intraosseous access and return of spontaneous circulation during out of hospital cardiac arrest. American Journal of Emergency Medicine, 35(2), 222-226. doi:10.1016/j.ajem.2016.10.052
  6. Burgert, J. M., Austin, P. N., & Johnson, A. (2014). An evidence-based review of epinephrine administered via the intraosseous route in animal models of cardiac arrest. Military Medicine, 179(1), 99-104. doi:10.7205/MILMED-D-13-00231
  7. Voelckel, W. G., Lurie, K. G., McKnite, S., Zielinski, T., Lindstrom, P., Peterson, C., Wenzel, V., & Lindner, K. H. (2001). Comparison of epinephrine with vasopressin on bone marrow blood flow in an animal model of hypovolemic shock and subsequent cardiac arrest. Critical Care Medicine, 29(8), 1587-1592. doi:10.1097/00003246-200108000-00015
  8. Niemann, J. T., Rosborough, J., Hausknecht, M., Garner, D., & Criley, J. M. (1981). Blood flow without cardiac compression during closed chest CPR. Critical Care Medicine, 9(5), 380–381.
  9. Dalsey, W. C., Barsan, W. G., Joyce, S. M., Hedges, J. R., & Lukes, S. J. (1984). Comparison of superior vena caval and inferior vena caval access using a radioisotope technique during normal perfusion and cardiopulmonary resuscitation. Annals of Emergency Medicine, 13(10), 881–884. doi:10.1016/S0196-0644(84)80661-7
  10. Pasley, J., Miller, C. H., DuBose, J. J., Shackelford, S. A., Fang, R., Boswell, K., Halcome, C., Casey, J., Cotter, M., Matsuura, M., Relph, N., Tarmey, N. T., & Stein, D. M. (2015). Intraosseous infusion rates under high pressure: A cadaveric comparison of anatomic sites. Journal of Trauma and Acute Care Surgery, 78(2), 295-299. doi:10.1097/TA.0000000000000516
  11. Hoskins, S. L., do Nascimento, P. Jr., Lima, R. M., Espana-Tenorio, J. M., & Kramer, G. C. (2012). Pharmacokinetics of intraosseus and central venous drug delivery during cardiopulmonary resuscitation. Resuscitation, 83(1), 107-112. doi:10.1016/j.resuscitation.2011.07.041
  12. Ong, M. E., Chan, Y. H., Oh, J. J., & Ngo, A. S. (2009). An observational, prospective study comparing tibial and humeral intraosseous access using the EZ-IO. American Journal of Emergency Medicine, 27(1), 8-15. doi:10.1016/j.ajem.2008.01.025
  13. Donnino, M. W., Salciccioli, J. D., Howell, M. D., Cocchi, M. N., Giberson, B., Berg, K., Gautam, S., & Callaway, C. (2014). Time to administration of epinephrine and outcome after in-hospital cardiac arrest with non-shockable rhythms: Retrospective analysis of large in-hospital data registry. BMJ, 348, g3028. doi:10.1136/bmj.g3028
  14. Andersen, L. W., Kurth, T., Chase, M., Berg, K. M., Cocchi, M. N., Callaway, C., & Donnino, M. W. (2016). Early administration of epinephrine (adrenaline) in patients with cardiac arrest with initial shockable rhythm in hospital: Propensity score matched analysis. BMJ, 353, i1577. doi:10.1136/bmj.i1577