Background
Intraosseous (IO) access is an incredibly useful skill in prehospital emergency care, especially when time is of the essence and intravenous access proves challenging. I have, on occasions, performed anaesthesia and blood transfusion only using IO needles prehospitally in situations that felt very challenging, but without it, there would have been few other options (alongside subclavian access as another great tool). IO is often the first choice in cardiac arrest, paediatrics and in some trauma patients.
It’s also incredibly useful for the patient you managed to get one IV line into, but that second one is proving hard to get. But as with any medical procedure, it’s not without its potential risks. Questions have been raised about whether IO use might lead to long-term complications such as infections like osteomyelitis, bone necrosis (osteonecrosis), or even compartment syndrome. Most of the data so far have been based on small case series or anecdotal reports. This week we have a study from Denmark that provides a clearer picture, looking at the issue systematically using data from national health registries. The abstract is below, but as always, we recommend that you read the full paper yourself and come to your own conclusions.
Abstract
Background: The guidelines for Advanced Life Support issued by the European Resuscitation Council recommend considering drug delivery through intraosseous access if intravenous access to the vascular bed is not feasible or unsuccessful. Emergency prehospital intraosseous cannulation may theoretically lead to an increased risk of long- term complications such as osteomyelitis, osteonecrosis, or compartment syndrome. Such complications have previously been reported in case reports or small sample case series. We systematically investigated long-term complications potentially associated with intraosseous cannulation using validated Danish health registries.
Methods: Data sources were the nationwide electronic Prehospital Patient Record system, the Danish National Patient Registry, and the Danish Civil Personal Registry. We investigated all patients who were subjected to prehospital intraosseous cannulation in Denmark from January 2016 through December 2019. During a follow- up period of 180 days from the index date we extracted information concerning mortality status and potential long-term complications defined as osteomyelitis, osteonecrosis, or compartment syndrome from the day of prehospital intraosseous cannulation.
Results: Of the 5,387 patients receiving intraosseous access, 375 were unidentified and lost to follow-up. Of the 5012 remaining patients, 4,775 were adults, and 237 were children. No children and “less than five” adults had long-term complications. No osteonecrosis, osteomyelitis or compartment syndrome appeared later than 175 days after an intraosseous cannulation.
Louise Breum Petersen, Søren Bie Bogh, Peter Martin Hansen, Louise Milling, Jens Stubager Knudsen, Helena Pedersen, Erika F. Christensen, Ulla Væggemose, Fredrik Folke, Signe Amalie Wolthers, Helle Collatz Christensen, Anne Craveiro Brøchner, Søren Mikkelsen, An assessment of long-term complications following prehospital intraosseous access: A nationwide study, Resuscitation, Volume 206, 2025, 110454, ISSN 0300-9572, https://doi.org/10.1016/j.resuscitation.2024.110454.
What kind of study is this?
This research is a retrospective, register-based cohort study leveraging data from three Danish national databases. The Electronic Prehospital Patient Record (ePPR) was used to identify instances of IO cannulation, the Danish National Patient Registry (DNPR) to track hospital diagnoses of long-term complications, and the Civil Personal Register to determine mortality and follow-up outcomes. Patients treated prehospitally with IO access between January 2016 and December 2019 were included. By utilising unique personal identification numbers, researchers followed patients for 180 days post-procedure, examining their outcomes for predefined complications. What makes this paper different from others in this area is the comprehensive and coordinated way they have followed up the patients using established national datasets.
Tell me about the patients
The study initially included 5,387 patients who underwent prehospital IO cannulation. After excluding 375 patients due to incomplete follow-up data, the final cohort comprised 5,012 patients, including 4,775 adults and 237 children. Most patients were aged over 60 years, with children under 18 forming a smaller subgroup, predominantly aged 0-2 years. The gender distribution was balanced, with a slight predominance of males (60.6%).
What were the measured outcomes in this study?
The primary outcomes were the diagnosis of long-term complications, including osteomyelitis, osteonecrosis, and compartment syndrome. Secondary outcomes included mortality at 180 days, demographic details such as age and gender, and diagnoses associated with IO access based on ICD-10 classifications.
What are the main results?
- Complications: Fewer than five cases (<0.1%) of osteomyelitis were recorded in adults. No cases of osteonecrosis or compartment syndrome were identified. No long-term complications were observed in children.
- Mortality: The overall mortality at 180 days was 60%. Mortality was highest in patients over 60 years old, reaching 68.8%.
- Common diagnoses: Most frequent diagnoses fell under circulatory system diseases, such as cardiac arrests, and injuries.
- Drug administration: Adrenaline was the most commonly delivered medication, consistent with its use in resuscitation scenarios.
Should we believe the results?
This study has several strengths. Its nationwide scope and inclusion of nearly all prehospital intraosseous uses in Denmark appears to be pretty unique, assuming that the data quality is good. We can’t really test that here and have to assume that it is. It’s also good to see robust data linkage using unique personal identifiers across databases enabled comprehensive tracking of outcomes, this means that the effective follow up rate is very high. I also like the use of predefined outcome measures based on ICD-10 codes, ensuring consistency in reporting.
However, no study is perfect, and there are elements we need to consider. Patients without valid identification were excluded, potentially skewing results and the follow-up period of 180 days may have underestimated complications like osteonecrosis, which could manifest later. We must also be mindful that many of the patients died and died early, and so did not have the opportunity to develop complications. The study population’s high mortality rate reflects the severity of conditions necessitating IO access, but that does mean that 50% were dead within a few days, and so would not have had any chance of developing a long-term complication. One way to look at this might be to assume that the same proportion of patients in the cohort who died might have developed complications at the same rate as those who survived. As it’s roughly half and half, perhaps the true complication rate of the procedure is twice that reported in the headline data? That’s a huge presumption and guess, but hopefully, you follow the logic.
Lastly, reliance on registry accuracy introduces the possibility of misclassification or underreporting.
Should we change practice based on this study?
This study supports the safety of IO access in prehospital settings, showing a low risk of long-term complications. It reinforces confidence in intraosseous usage for scenarios requiring rapid vascular access, such as cardiac arrests and major trauma and I will still be continuing to use them. It would be great to see some even longer-term outcome data and perhaps to look at different sites. However, the low event rate here means that we would need huge numbers and/or a reliance on case series reports to investigate this.
Summary
This large-scale Danish study provides good evidence for the safety of prehospital IO access, with fewer than 0.1% of patients experiencing long-term complications. I feel reassured to continue my practice of using this technique in prehospital and emergency care.
vb
S
References
- Louise Breum Petersen, Søren Bie Bogh, Peter Martin Hansen, Louise Milling, Jens Stubager Knudsen, Helena Pedersen, Erika F. Christensen, Ulla Væggemose, Fredrik Folke, Signe Amalie Wolthers, Helle Collatz Christensen, Anne Craveiro Brøchner, Søren Mikkelsen, An assessment of long-term complications following prehospital intraosseous access: A nationwide study, Resuscitation, Volume 206, 2025, 110454, ISSN 0300-9572, https://doi.org/10.1016/j.resuscitation.2024.110454.
- A. Granfeldt, S.R. Avis, P.C. Lind, et al. Intravenous vs. intraosseous administration of drugs during cardiac arrest: a systematic review Resuscitation, 149 (2020), pp. 150-157, 10.1016/j.resuscitation.2020.02.025
- J. Soar, B.W. Böttiger, P. Carli, et al. European resuscitation council guidelines (2021): adult advanced life support. Resuscitation, 161 (2021), pp. 115-151, 10.1016/j.resuscitation.2021.02.010
- Tyler J, Perkins Z, De’Ath H, Smith IM, Brohi K. Intraosseous access in the resuscitation of trauma patients: a literature review. Eur J Trauma Emerg Surg. 2016;42(2):213-223. doi:10.1007/s00068-020-01327-y
- Intraosseous access in the resuscitation of patients with trauma: the good, the bad, the future. Qasim ZA, Joseph B.Trauma Surg Acute Care Open. 2024 Apr 15;9(Suppl 2):e001369. DOI: 10.1136/tsaco-2024-001369
