Bone infusion, formally called intraosseous (IO) infusion, is a method of delivering fluids and medications directly into the marrow cavity of a bone, where they are rapidly absorbed into the bloodstream. The technique works because the spongy interior of bones contains a dense network of blood vessels that do not collapse during shock or cardiac arrest, making them a reliable pathway when standard veins are inaccessible. First developed just before World War II, bone infusion fell out of routine adult use for decades before being revived as a cornerstone of modern emergency medicine, military trauma care, and critical resuscitation.
How Bone Infusion Actually Works
The inner cavity of most large bones is filled with marrow, a soft tissue laced with tiny blood vessels called sinusoids. These sinusoids drain into larger venous channels that feed directly into the central circulation. When a needle is placed through the hard outer shell of a bone and into the marrow space, anything pushed through that needle enters those sinusoids and reaches the heart and the rest of the body within seconds to minutes. The mechanism is essentially the same as a standard intravenous (IV) line, just using the bone’s internal vascular network instead of a peripheral vein.
What makes this route so valuable in emergencies is that peripheral veins tend to collapse when blood pressure drops severely, as in major bleeding or cardiac arrest. The bony vessels, protected inside rigid walls, stay open even when a patient is in profound shock. That reliability is the entire reason the technique exists: it is a backup vascular highway that stays functional when the usual roads shut down.
A Brief and Surprising History
The discovery of intraosseous infusion was something of an accident. Just before the outbreak of World War II, researchers stumbled on the idea that fluids and drugs could be pushed through bone marrow and reach the bloodstream efficiently.1PubMed Central. Discovery of the intraosseous route for fluid administration The technique was used in both adults and children during the war years. After that, though, it largely disappeared from adult emergency medicine for decades. Pediatric medicine kept it alive, since tiny children often have veins that are nearly impossible to access, but adult resuscitation protocols moved on to other options. Interest in adult IO access resurged in the 1980s and 1990s as new devices made insertion faster and emergency physicians recognized the need for a reliable vascular route when peripheral IVs fail.
When Bone Infusion Is Used
Current resuscitation guidelines position the IV route as the first choice for delivering drugs during emergencies, with the IO route reserved for patients in whom IV access cannot be quickly established.2Resuscitation. Drug routes in out-of-hospital cardiac arrest: A summary of current evidence In practice, this means bone infusion shows up most often in cardiac arrest, severe trauma with hemorrhagic shock, burns covering large areas of skin, severe dehydration in small children, and any situation where multiple attempts at a peripheral IV have failed. A systematic review of IO use in adults concluded that IO infusion should be implemented in all critical situations when peripheral venous access is not easily obtainable.3PubMed Central. Use of intra-osseous access in adults: a systematic review
The emphasis on “not easily obtainable” matters. Bone infusion is not a first-line choice for routine hospital care. It is a rescue route, and the decision to use it typically comes after one or two failed IV attempts, or immediately in situations like cardiac arrest where every second counts and the team cannot afford delays.
Where the Needle Goes
Three main anatomic sites are used for IO insertion, each with different strengths.
The proximal tibia, the flat area just below the kneecap on the inner side of the shinbone, is the most commonly taught and most widely used site. It is easy to find by feel, even in obese patients, and works well in both adults and children. In cadaver studies, tibial IO access delivered a mean flow rate of about 31 mL per minute under high pressure.4Journal of Trauma and Acute Care Surgery. Intraosseous infusion rates under high pressure: A cadaveric comparison of anatomic sites
The proximal humerus, the rounded top of the upper arm bone near the shoulder, offers a site closer to the central circulation. Flow rates there averaged about 57 mL per minute, roughly double the tibial rate.4Journal of Trauma and Acute Care Surgery. Intraosseous infusion rates under high pressure: A cadaveric comparison of anatomic sites The humerus is particularly useful in military and tactical settings, where lower extremity injuries may make the tibia unavailable.
The sternum, or breastbone, consistently produces the highest flow rates. In cadaver testing, sternal IO achieved a mean flow rate of about 94 mL per minute, roughly 1.6 times the humeral rate and 3.1 times the tibial rate.4Journal of Trauma and Acute Care Surgery. Intraosseous infusion rates under high pressure: A cadaveric comparison of anatomic sites Another review confirmed that flow was good or very good in nearly 90% of sternal insertions.5PubMed Central. Sternal Intraosseous Devices: Review of the Literature – Section: Flow Rates The sternal site requires a specialized device and carries the serious risk of mediastinal injury if the needle goes too deep, so it is used more selectively than the tibia or humerus.
Modern Devices and How Insertion Works
Early IO needles were inserted by hand, twisting a thick-gauge needle through the bone cortex with manual force. This worked, but it was slow, physically demanding, and had a meaningful failure rate. In one study, manual IO insertion in adult simulation models succeeded only three out of 22 times, compared with a 100% success rate for a battery-powered drill device.6PubMed Central. Comparison of mechanical and manual bone marrow puncture needle for intraosseous access; a randomized simulation trial Insertion times also dropped dramatically: adult IO insertion with a drill took about 3.7 seconds versus nearly 55 seconds manually.6PubMed Central. Comparison of mechanical and manual bone marrow puncture needle for intraosseous access; a randomized simulation trial
The most widely used modern device is a battery-powered drill (the EZ-IO is the brand you will encounter most often). In animal models, drill-assisted placement averaged under four seconds with a 100% success rate, while manual placement averaged over 33 seconds with a 76% success rate, and the most common manual complication was a bent needle.7PubMed Central. Effectiveness of a Drill-assisted Intraosseous Catheter versus Manual Intraosseous Catheter by Resident Physicians in a Swine Model In human use, the EZ-IO achieved successful first-attempt access about 98% of the time, compared with about 80% for manual needles, and had zero technical complications versus about 15% for the manual approach.8Resuscitation. Comparison of two intraosseous infusion systems for adult emergency medical use
In pediatric simulations, both manual and mechanical needles achieved 100% success in child and infant models, though the drill was still significantly faster.6PubMed Central. Comparison of mechanical and manual bone marrow puncture needle for intraosseous access; a randomized simulation trial Children’s bones are softer and thinner, which explains why manual insertion is more feasible in younger patients and why IO access has long been a pediatric staple.
Do Drugs Work the Same Through Bone?
One of the most important questions about bone infusion is whether medications reach the bloodstream as effectively as they would through a regular IV. The answer depends on which bone you are infusing into.
For the sternal route, the pharmacokinetics are remarkably similar to central venous access. A study comparing sternal IO with central venous delivery during cardiac arrest found that the time to peak blood concentration was not significantly different, and the total dose delivered through the sternum was about 86% of the central venous dose.9Resuscitation. Pharmacokinetics of intraosseous and central venous drug delivery during cardiopulmonary resuscitation Tibial IO, however, was slower: peak arterial drug levels took about twice as long to appear (107 seconds versus 53 seconds for sternal), and the total dose delivered was only 65% of the sternal amount.9Resuscitation. Pharmacokinetics of intraosseous and central venous drug delivery during cardiopulmonary resuscitation
For pain medications specifically, a pharmacokinetic study of morphine found no significant differences between IO and IV administration for nearly all measured parameters, including peak concentration, time to peak, and total drug exposure. The researchers concluded the two routes were bioequivalent for morphine in adults.10American Journal of Emergency Medicine. Does intraosseous equal intravenous? A pharmacokinetic study This finding matters well beyond morphine: it supports the broader principle that nearly any drug that can be given intravenously can also be given through bone marrow.
Bone Infusion in Cardiac Arrest
Cardiac arrest is where bone infusion gets its highest-profile use, and where the biggest clinical question has played out: does IO access produce outcomes as good as IV access? For years, observational data painted a murky picture, with some studies suggesting IO patients did worse. The problem was confounding: patients who get IO access are often sicker or harder to treat, which makes the IO route look bad by association.
A major randomized trial published in the New England Journal of Medicine directly compared initial IO versus initial IV access in adults with out-of-hospital cardiac arrest. Sustained return of spontaneous circulation (meaning the heart started beating again on its own) occurred in about 30% of the IO group and 29% of the IV group, with no significant difference between them.11PubMed. Intraosseous or Intravenous Vascular Access for Out-of-Hospital Cardiac Arrest A separate systematic review and meta-analysis pooling multiple randomized trials also found no significant differences for 30-day survival, survival to hospital discharge, or favorable neurological outcomes.12PubMed. Intravenous vs intraosseous administration of drugs for out of hospital cardiac arrest: A systematic review and meta-analysis A third meta-analysis of randomized controlled trials confirmed the same pattern: a numerically small difference in return-of-circulation rates that did not reach statistical significance.13PubMed Central. Intraosseous versus intravenous vascular access in out-of-hospital cardiac arrest: a systematic review and meta-analysis of randomized controlled trials
The takeaway from this collective evidence is reassuring: when a patient needs drugs during cardiac arrest and you cannot get an IV quickly, the IO route does not compromise their chances. It is a genuine equivalent in the setting that matters most.
Complications and Safety Profile
Bone infusion complications are uncommon but real, and they fall into a few categories.
Extravasation, where fluid leaks out of the bone and pools in the surrounding soft tissue, is the most clinically concerning risk. Reported cases include severe tissue necrosis in an infant and compartment syndrome requiring surgical release in a near-drowning victim.14Annals of Emergency Medicine. Intraosseous Extravasation Complication Reports Extravasation typically happens when the needle misses the marrow cavity, passes through the bone completely, or when infusion pressure is too high. Monitoring for swelling around the insertion site is the primary way clinicians catch this early.
Fat and bone marrow emboli are a theoretical concern that has been extensively studied. When you push fluid through bone marrow, small globules of fat and marrow particles inevitably get swept into the bloodstream and travel to the lungs. Animal studies consistently found these microemboli in lung tissue after IO infusions, but they did not cause any measurable change in lung function during four hours of observation. The researchers concluded that although fat and bone marrow emboli are essentially universal after IO drug administration, they are not of immediate clinical importance and should not discourage use of the IO route when it is needed.15Annals of Emergency Medicine. The safety of intraosseous infusions: Risks of fat and bone marrow emboli to the lungs Another study using a CPR model found that IO infusion did not increase fat embolization over what CPR alone produced.16The American Journal of the Medical Sciences. Fat Embolism With the Use of Intraosseous Infusion During Cardiopulmonary Resuscitation More recent work has shown that subclinical pulmonary fat emboli can persist 24 hours after IO infusion and that fat intravasation increases with higher infusion rates.17PubMed. Fat Intravasation from Intraosseous Flush and Infusion Procedures The clinical significance of these persistent emboli in surviving patients is still not fully understood.
Other potential complications include infection at the insertion site, needle dislodgement, and pain during infusion (bone marrow is not numb, so a flush of lidocaine is often given first in conscious patients). Osteomyelitis, or bone infection, has been reported but is rare when the needle is removed within 24 hours, which is the standard recommendation.
Can You Draw Blood From Bone?
An underappreciated feature of IO access is that you can aspirate blood samples from the marrow cavity before you start infusing fluids, and those samples can substitute for venous blood in some laboratory tests. This matters because the same patients who need IO access are the same patients who desperately need lab values, and they often have no veins to draw from.
IO blood samples correlate well with venous blood for hemoglobin, hematocrit, and most of a basic chemistry panel including glucose, blood urea nitrogen, creatinine, and chloride. They do not reliably match for white blood cell counts, platelet counts, or electrolytes like potassium and calcium.18PubMed. A new study of intraosseous blood for laboratory analysis Point-of-care analysis has found good agreement between IO and arterial blood for pH, glucose, and lactate, though potassium is consistently higher in IO samples and hemoglobin values show wide variation.19PubMed Central. Intraosseous blood samples for point-of-care analysis: agreement between intraosseous and arterial analyses Another analysis confirmed that IO and IV samples were most comparable for pH, bicarbonate, sodium, and base excess.20The American Journal of Emergency Medicine. Analysis of intraosseous blood samples using an EPOC point of care analyzer during resuscitation
In short, IO blood is usable but not identical to venous blood. Clinicians who understand which values are reliable and which are skewed can still extract critical diagnostic information from a bone marrow aspirate during resuscitation.
Bone Infusion in Children
Pediatric emergency medicine was the specialty that kept IO access alive when adult medicine largely forgot about it. Children, especially infants, have small veins that constrict rapidly in shock, making IV placement extraordinarily difficult in the sickest patients. The proximal tibia in a child is close to the skin surface and easy to access, which is why it became the go-to site for pediatric IO insertion decades ago.
A reasonable concern with children is whether driving a needle through a growing bone might damage the growth plate and affect limb development. An animal study in young pigs found no growth disturbances or growth plate abnormalities clinically or on X-ray through six months of rapid skeletal growth after IO infusion through the tibial growth plate. The researchers concluded that IO infusion was safe with no significant growth defects despite the mechanical injury of needle placement.21ScienceDirect / Annals of Emergency Medicine. Analysis of growth plate abnormalities following intraosseous infusion through the proximal tibial epiphysis in pigs While animal data always have limits, this study has been reassuring enough that pediatric IO use is considered standard practice in emergencies worldwide.
Military and Tactical Applications
The modern revival of IO access in adults owes a lot to military medicine. Combat injuries, particularly from explosions, often destroy the very limbs where you would normally start an IV. In Afghanistan, UK military medical teams used IO access on 26 patients over three months of combat casualty care, inserting 32 needles with 97% effective function. Fluid including blood products and drugs ranging from painkillers to cardiac arrest medications to anesthetics were delivered successfully through bone.22PubMed. Intra-Osseous Access (EZ-IO) for Resuscitation: UK Military Combat Experience
The humeral head (top of the upper arm bone near the shoulder) has become especially important in tactical settings because a casualty wearing body armor can have the arm exposed while the legs are mangled. A cadaver feasibility study of the humeral head approach achieved 96% first-pass success with a drill device and 100% success by the second attempt.23PubMed. A review of the evolution of intraosseous access in tactical settings and a feasibility study of a human cadaver model for a humeral head approach IO devices are now standard equipment in many military first-aid kits, because they are small, light, and can be used by medics with relatively brief training under conditions where starting an IV would be impractical.
Training and Skill Retention
One of IO access’s practical advantages is how quickly the skill can be learned, at least when using modern drill devices. A simulation-based training study found that resident physicians’ confidence in IO placement increased significantly after a training session, and both confidence and knowledge remained elevated at a two-month follow-up. Knowledge scores improved by about 38% immediately after training and stayed about 22% higher than baseline at two months.24Annals of Emergency Medicine. Simulation-Based Training for Intraosseous Line Placement in Pediatric Emergency Medicine
Skill retention is not perfect, however. A randomized trial comparing two IO devices among hospital providers tested knowledge and practical skills at 3 and 12 months after initial training. While theoretical knowledge held up over time, practical execution revealed gaps: common errors included skipping site disinfection and improper attachment of the needle to the IV tubing. In about a third of handlings with one particular device, unsafe practices were observed.25PubMed. A randomized trial comparing two intraosseous access devices in intrahospital healthcare providers with a focus on retention of knowledge, skill, and self-efficacy The evidence suggests that learning IO insertion is straightforward, but maintaining competency requires periodic refresher practice, especially since the procedure is used infrequently by most clinicians.
Cost Compared to Central Lines
When a peripheral IV fails and IO is not used, the usual alternative in hospitals is a central venous catheter (CVC), a longer line threaded into a large vein in the neck, chest, or groin. Central lines are effective but time-consuming, require sterile technique, and carry their own complications including pneumothorax and bloodstream infections. A cost analysis found that for straightforward placements, IO access saved about $195 per procedure compared to a central line. The authors estimated that if just 20% of the roughly 3.5 million CVCs placed annually in the United States were replaced with IO catheters, savings could approach $650 million per year.26PubMed. Intraosseous vascular access is safe, effective and costs less than central venous catheters for patients in the hospital setting Speed was another factor: IO catheters were placed faster with higher success rates and fewer complications than CVCs in that analysis.
This economic argument has not yet led to IO access replacing central lines on a large scale in hospital wards, partly because IO needles are designed for short-term use (hours, not days) and partly because clinician habits favor familiar procedures. But in emergency departments and during resuscitation, the speed and cost advantages of bone infusion are increasingly recognized as practical benefits on top of the clinical ones.