The Way to a Patient’s Heart — Vascular Access in Cardiac Arrest

Niklas Nielsen

N Engl J Med Published October 31, 2024

DOI: 10.1056/NEJMe2412901

Every year, thousands of lives are saved with cardiopulmonary resuscitation (CPR). Yet, overall survival among patients with out-of-hospital cardiac arrest is approximately 10%, even in well-performing emergency medical systems. The interventions that are undertaken at the scene of a cardiac arrest are the most likely interventions to improve outcomes: early alert, chest compressions, defibrillation, and ventilation after collapse.¹

Vascular access to facilitate drug delivery has been an integral part of CPR for decades because the administration of epinephrine increases the chances of return of circulation, improves the likelihood of survival to the time of hospital admission, and is recommended in guidelines.^2,3 Whether the use of epinephrine during cardiac arrest leads to more patients returning home with a good outcome is less clear, and the available evidence for the use of drugs other than epinephrine is even less certain. To evaluate these interventions in clinical trials, an effective route to the vasculature is essential.

Intraosseous vascular access, in which a cannula is placed in the spongy part of a bone, is an alternative to intravenous access and was proposed as an entry point to the systemic circulation 100 years ago.⁴ Intraosseous-access systems have since evolved and offer a procedure that is fast to perform and easy to learn. With different systems available, the success rate of a first attempt at intraosseous access is higher than that of a first attempt at peripheral venous access.⁵ However, intraosseous access has several possible downsides, such as a potentially reduced effect of the drugs injected, a lower infusion speed, risk of infection, and contraindications in patients with fractures or protheses.⁶ Intraosseous access is also associated with a higher cost than intravenous access.⁷ The American Heart Association and the European Resuscitation Council recommend a strategy of attempting to establish intravenous access first, followed by the intraosseous approach if the attempt at intravenous access is unsuccessful.^2,3

In the past several years, the use of intraosseous access has increased,⁸ while observational data and the results of a cluster-randomized trial have indicated neutral or worse outcomes with intraosseous access.^7,9,10 Accordingly, there have been calls for randomized clinical trials in which first attempts at intravenous vascular access are compared with first attempts at intraosseous vascular access to determine which strategy is better.⁶

The results of two independent, randomized clinical trials — which together included more than 7500 patients with out-of-hospital cardiac arrest — are now reported in the Journal. Both the PARAMEDIC-3 trial¹¹ (conducted in the United Kingdom) and the Intravenous vs. Intraosseous Vascular Access during Out-of-Hospital Cardiac Arrest (IVIO) trial¹² (conducted in Denmark) compared a strategy of intraosseous access as a first attempt with a strategy of intravenous access as a first attempt.

Although the primary outcomes in the two trials differed (survival at 30 days in the PARAMEDIC-3 trial; sustained return of spontaneous circulation in the IVIO trial), the results of the trials were remarkably consistent and complementary. Each was performed according to the highest methodologic standards, and each had a low risk of bias, except for the limitation that the randomization to intraosseous access or intravenous access was inherently impossible to mask for the emergency medical crews and in-hospital personnel. The standardized nature of CPR and the strict algorithms for termination of resuscitation would minimize any bias from a primarily unblinded intervention. The IVIO trial included a blinded assessment of neurologic outcomes, the results of which supported those of the unblinded primary outcome analysis. A limitation of the PARAMEDIC-3 trial was that recruitment was stopped early, when less than half the ambitious planned sample size had been reached; however, the analyses indicated that continuation toward the planned sample size would have had little influence on the final results.

The findings and designs of the trials complement each other. The PARAMEDIC-3 trial was larger than the IVIO trial and involved a more pragmatic approach. It was designed with a patient-centered primary outcome: survival at 30 days. Although the primary outcome in the IVIO trial was less patient-centered (sustained return of spontaneous circulation), it enabled a smaller sample size. Neither trial showed a benefit with respect to the primary outcomes between the intraosseous-first and intravenous-first approaches, and these results were supported by the results of the subgroup and sensitivity analyses. No apparent difference was seen between the two vascular-access approaches in the incidence of adverse events, which were generally uncommon. The results of the PARAMEDIC-3 trial suggest the possibility that patients who undergo intravenous access may be more likely to have a sustained return of spontaneous circulation than those who undergo intraosseous access; such a finding is important because this outcome is a prerequisite for any further favorable outcome for patients.

The IVIO trial provides further detailed data on the quality of CPR, which is especially worth highlighting. The two approaches to vascular access did not lead to marked differences in procedural variables such as the depth and rate of chest compressions or the time to the initial receipt of epinephrine. Of note, the median time to drug delivery was 9 minutes longer in the PARAMEDIC-3 trial than in the IVIO trial (24 minutes vs. 15 minutes), a finding that may be related to the striking difference in overall survival between the trials (with survival being lower in the PARAMEDIC-3 trial) and should lead to systematic evaluation of the way that emergency medical services are organized. On the other hand, this result indicates that any difference between the outcomes with intraosseous access and intravenous access is probably not time-dependent.

The PARAMEDIC-3 and IVIO trials do not support a change in the current guideline recommendation to use an intravenous-first strategy, with intraosseous access as a secondary approach, but these trials move the certainty of evidence from very low toward high. An independent, patient-level meta-analysis and an analysis of health economics and resource consumption would further inform guidelines and practice.