Design, Development, and Validation of an Intra-Osseous Needle Placement Guide

Fluid resuscitation is a fundamental tenet of treating patients in shock [1], and the association of shock with trauma has long been recognized. There are four types of shock: (1) septic, (2) hypovolemic (or hemorrhagic), (3) neurogenic, and (4) cardiogenic [2]. They are distinct, each with different etiologies but with similar initial treatment strategies. Septic shock is attributable to infection with bacteria or viruses [3]. Hypovolemic shock is characterized by inadequate intravascular volume resulting from hemorrhage, dehydration, or as a result of burns or third spacing (occurs when intravascular oncotic pressure is decreased) [4]. Shock caused by trauma results from severe tissue damage, whether in the form of blunt trauma causing soft tissue injury, blood loss from multiple fractures, or burns [5]. True cardiogenic shock results from myocardial damage resulting in low cardiac output, hypotension, and clinical signs of inadequate tissue perfusion [2].

Although these four types of shock are separate, they often overlap and intravenous (IV) fluid resuscitation is fundamental for managing the unstable patient [5]. The most common types of shock encountered in an acute setting (in both the developed and developing world) result from trauma, sepsis, or causes of hypovolemia like postpartum hemorrhage or dehydration [2]. Administration of large volumes of intravenous fluids requires intravascular access [6], achieved through central lines or large-bore IV in a hospital setting. However, in the prehospital or remote setting it could be achieved via an IV or an intra-osseous (IO) line [7]; placement of these two access strategies is depicted in Fig. 1.

Placement of an IV line requires skilled and trained healthcare providers and requires multiple peripheral sites in the extremities; the upper extremities are more often used based on convention and ease of maintenance. Moreover, even with appropriate training, providers may not reliably be able to place an IV due to severe hypovolemia causing venous collapse [8]. In these cases, placement of an IO catheter would be significantly faster and much less painful for the patient, than multiple sticks and attempts to place an IV [8–10].

Intra-osseous lines have already been adopted by many emergency medical services in the U.S. as well as the U.S. military [11–15]. IO infusion works through delivery of fluid to highly vascularized bone marrow, which is in turn drained by medullary venous channels and nutrient and emissary veins. The benefit of infusion into the intramedullary cavity is that it is a noncollapsible system, even in the presence of profound hypovolemia [16]. The fluid delivery rate through an IO is comparable to an IV catheter. Most drugs or medications normally given through an IV can also be administered through an IO catheter and with the same efficacy as well [17,18]. While motivated by the limitations associated with delivering medical care to those in the developing world, we sought to develop a device that would be low cost and easy to use, so that it would be easily deployed in a wide range of situations and therefore help the maximum number of patients.

The most common sites for insertion of an IO catheter are as follows: (1) proximal tibia, (2) proximal humerus, and (3) sternum [19]. Figure 2 depicts the three different sites for placement of IO catheters. Both the proximal tibia [20] and proximal humerus [21] have high insertion success rates and are easy and quick sites to place an IO line with minimal training. There is increased risk associated with sternal IO placement, as a too-long or malaligned needle can cause a hemothorax or pneumothorax [22].

Further, sternal IO interferes with effective delivery of cardiopulmonary resuscitation [23]; as such, tibial and humeral IO are the preferred sites. Wampler et al. [24] suggested that the proximal humerus would be more difficult to access than the tibia not as a result of finding landmarks [7,17,25], but more related to the close proximity of this site to the central circulation. The proximal tibia is the preferred insertion site as it has easily identifiable landmarks [26], and the underlying bone is relatively superficial even in obese patients, with only a thin layer of skin and subcutaneous connective tissue [27]. The tibia at this location has a relatively thin cortex, presents a flat and wide target for healthcare providers [28,29], and poses minimal morbidity for malaligned insertions [30]. Further, one study from 1985 demonstrated an extremely low infection rate of 0.6% while monitoring 4270 IO cannulations in the pediatric population [31].

The Tib-Finder is a drill guide to be used to ensure accurate placement of IO catheters with minimal patient contact by an inexperienced provider and to improve the ease-of-use of such devices by experienced, trained providers. Using the Tib-Finder, an IO line can easily be placed in a patient who otherwise would be at risk for dying from shock; fluids can be administered in the field by first responders prior to triage, or in cases of delayed triage due to a large number of patients they can help sustain patients until transport. It is a one-size-fits-all device designed based on anthropometric measurements, with demonstrated flexibility for use with a wide range of patient sizes, ranging from children to adults.

The Tib-Finder was reduced to practice as the result of a rapid, 90-day design cycle; its creation was initially motivated by the need for an improved emergency response to the recent Ebola epidemic in West Africa [32]. A deterministic, iterative design process was followed; the design team utilized anthropometric data gathered from human skeletons to guide the sizing and placement of ports and also performed studies of unskilled individuals' ability to execute simple tasks like drilling and centering to inform decisions like handle placement and indicator markings. A design freeze was instituted in the Fall of 2016 and the current Tib-Finder design is moving toward production and distribution.

The first Tib-Finder design concepts were based on a review of IO needle placement techniques and discussions with healthcare workers who had experience treating patients in West Africa during the Ebola crisis. Functional requirements for the Tib-Finder are as follows:

1. (1)

   locates IO drill in proper position and orientation in proximal tibia,

2. (2)

   exposes proximal medial tibia below joint line,

3. (3)

   flexibility to be used on a range of patient sizes (small child to large adult; underweight to obese),

4. (4)

   easy, intuitive use for care givers with wide range of experiences and also wide range of language backgrounds,

5. (5)

   eliminates need for direct contact with patient,

6. (6)

   functions on either side of the patient's body,

7. (7)

   able to be injection molded, little to no assembly required,

8. (8)

   durable––enables global distribution to remote locations, and

9. (9)

   does not interfere with normal use of IO drill.

Development of the tool began as a design exercise, performed over a 1-week period in October 2014. Team members interviewed physicians and practiced with an existing IO placement system and mannequins in a simulated patient environment. The “ice skate” design concept that was created in this first stage is shown in Fig. 3(a). The Tib-Finder initially contained a large distal slot for IO placement further down the tibia as a countermeasure if proximal placement failed. What followed was a 3-month period of concept development and iteration, based on the 48 h “Design, Prototype, Test, Criticize, Iterate” cycle [31]; during this 3-month period, anthropometric measurements and user performance data were gathered and incorporated into the design process to create additional concepts.

Figure 3(b) shows each Tib-Finder conceptual iteration, with the ice skate concept in the upper right and the final “shin guard” concept in the lower right. The alpha-prototype Tib-Finder can be seen in Fig. 3(c); this model is currently in a design-freeze while we work toward manufacturing Tib-Finders in accordance with FDA requirements. All the prototype Tib-Finders were cut out of acrylonitrile butadiene styrene thermoplastic polymer sheets on a small table-top milling machine (ProtoTRAK SMX, Southwestern Industries, Inc., Rancho Dominguez, CA); flat profiles of the base were first created in SolidWorks, and blanks were then thermoformed over an aluminum mold to ensure a consistent shape for all the prototypes. Handles were three-dimensional (3D) printed (Stratasys Mojo Desktop Printer, Eden Prairie, MN) and bonded each base using a flexible thermoplastic adhesive (Hardman, Double/Bubble Epoxy, Extra Fast Setting, South Bend, IN).

The design team identified the need to determine the location of a “safe zone” for IO needle placement, where a reasonable amount of misalignment would still result in the tip of the needle entering the intramedullar canal of the tibia. Three members of the team (including the lead authors A.H.S. and S.D.R.) traveled to the William M. Bass Donated Skeletal Collection at the Forensic Anthropology Center at the University of Tennessee, Knoxville, Knoxville, TN [33]. One hundred skeletal specimens were evaluated with a total of 195 tibiae measured; five skeletons had one tibia either missing or severely deformed due to congenital disease or other defect and so were not included in the analysis. The tibial tubercle was identified on all the specimens and used as a reference point for all the measurements; axes for defining measurement directions were medial–lateral, anterior–posterior, and proximal–distal (P–D); all the measurements were made using standard digital calipers (McMaster, Inc., Chicago, IL). The following measurements were made from the tubercle, as illustrated in Fig. 4 ¹: (1) P–D distance to oblique line, (2) P–D distance to anterior intercondylar region, (3) P–D distance to the inferior articular surface, (4) P–D distance to the soleal line (posterior), and (5) P–D distance to the nutrient foramen (posterior).

Next, starting at the tibial tubercle, a series of latitudes were defined, each further distal from the tubercle along the axis of the tibia. These began with a latitude of the tubercle (0), and then every 0.5 cm up to a total travel distance of 5 cm distal, yielding a total of 11 latitudes. At each latitude, the following were recorded: (1) the maximum thickness orthogonal to the medial surface; (2) the distance along the medial surface at a given latitude (roughly the AP direction, herein referred to simply as AP) from the anterior border (tibial spine) to the point of maximum thickness; (3) the total AP width of the medial surface; and (4) the thickness orthogonal to the medial surface 0.5 cm anterior and posterior to the point of maximum thickness.

Photographs of any notable abnormal tibiae were taken for review by the design team; they are not included here as they are outside the scope of this paper. A number of broken tibiae were also encountered; in these cases, the thickness of the cortex was measured when available and used to corroborate values provided by consulting clinicians; only a few broken tibia were encountered and their measurements were merely used as correlates for the design team when considering the size of each intra-osseous needle. Because the apex of the tibial tubercle is nearly always aligned with the distal aspect of the patella (as the tibial tubercle is the insertion of the patellar ligament), these measurements were deemed to be reliable and representative of those necessary to accurately locate a safe zone in the field based purely on the palpation technique. The location of the safe zone was defined as that area in the tibia, which would allow access to cancellous bone and placement of an IO catheter accurately with 99% certainty in 90% of individuals (within the 5th to 95th percentile of height).

The design team was unable to answer the question of how accurately an individual, whether a trained, skilled healthcare provider, or an unskilled layperson, would be able to drill a hole in the middle of a target region; further, the individual would need to ensure that the hole was orthogonal to the surface (i.e., the medial surface of the tibia) into which they were drilling. It was determined that a study would need to be performed to observe the behavior of unskilled human subjects in drilling holes without instruction. These data could then be utilized to guide any changes or adjustments to the Tib-Finder's design to mitigate the risk of malalignment of the IO drill. This study was approved by the Dartmouth College IRB prior to recruitment of subjects (Dartmouth CPHS Study 00028585).

A human factors skill study was designed to remove variability associated with anatomy and IO-specific technique. Rigid blocks of 20 lb/cubic foot test material (Last-A-Foam, General Plastics, Tacoma, WA) were machined into triangular cross sections. These were to be used to approximate the location of the medial surface of the tibia as presented to a caregiver (an approximate 45 deg angle). These blocks were secured to an elevated platform at which test subjects stood and were replaced after each subject's attempt. Tests were carried out on the Dartmouth College campus, and students who volunteered were enrolled to participate onsite; immediately after obtaining informed consent, study subjects were allowed to attempt the exercises so as to minimize their ability to prepare.

Upon providing consent, subjects (n = 48) answered a brief questionnaire about their background, including the following items:

1. (1)

   Whether their major was in one of the STEM disciplines?

2. (2)

   Whether they perceived themselves to be a “handy person”?

3. (3)

   Self-identifying as a “do-it-yourselfer.”

Subjects were then provided with a handheld, rechargeable drill with dimensions similar to that of the Arrow^® EZ-IO^® drill (Teleflex, Inc., Morrisville, NC) and asked to place five holes in a piece of paper mounted to the foam block. No instructions were provided with respect to where to place the holes or at what angle the holes should be drilled into the foam block. The only markings the subjects were given were five template circles on the paper (the circles were much larger than the diameter of the drill bit, requiring subjects to attempt to center the drill bit in the circular marking). After testing, the de-identified subject data were aggregated, and three points of interest were evaluated:

1. (1)

   a weighted composite score for mechanical ability (based on the questionnaire),

2. (2)

   the average distance that a hole was placed from the true center of the target by the subject, and

3. (3)

   the average angle away from vertical that a subject drilled while attempting to place the hole near the true center of the target.

The Tib-Finder also required a user guide, both as part of the design teams desire to create something that was user friendly and as part of the need to meet regulatory standards. It was anticipated that the Tib-Finder would be used in disaster situations worldwide and thus could not contain or utilize any languages; it would rely solely on pictorial schematics to illustrate to users how to safely place an IO line with the Tib-Finder. Specific requirements identified by the team were:

1. (1)

   no written words as part of user instructions,

2. (2)

   no assumptions made about a user's previous knowledge or experience,

3. (3)

   all the symbols are common/standard among cultures and societies, and

4. (4)

   multiple differentiating methods to delineate separate components or steps.

The design team drew inspiration from airline safety manuals as a model for generating the Tib-Finder instructions. Based on a written description of the process and referencing pictorial documentation, a graphic design firm (Funnel, Inc., Middleton, WI), with experience both in creating airline safety manuals and also with developing products for the medical field, created a draft pictographic user manual. An iterative approach was then used to refine the draft, incorporating a 24 h “Design, Test, Criticize, Iterate” cycle guided by the previously mentioned design criteria. To achieve this, a draft user manual was incorporated into an IRB-approved protocol with the design-frozen final Tib-Finder.

The study group was composed of 64 subjects, made up of physicians, nurses, EMTs, and other individuals both with and without technical backgrounds. Each day, five subjects at a time were invited to use the alignment guide and the EZ-IO system to place an IO line on a mannequin; a copy of the latest iteration of the user manual was also provided but no verbal instructions were given. Participants were all asked to place an IO needle into the proximal tibia of a mannequin using only the provided IO drill (Arrow^® EZ-IO^®, Teleflex, Inc., Morristown, NC), the Tib-Finder alignment guide, and the pictographic manual. Afterward, users were asked to complete a written feedback form, and based off of each day testing and feedback obtained from users, the manual was refined and improved. This cycle was repeated over the course of 2 weeks; the final version of the pictographic manual is shown in Fig. 5.

The Tib-Finder, a class 1 nonsignificant risk device, is currently in the final stages of production and validation, after which it will be available for use in human subjects. Its design is based on objective measurements of the human skeleton, and the ease-of-use was both refined and verified through human subjects testing on mannequins. The combination of objective data and subjective feedback from test subjects was utilized to create a device that makes placement of an IO line simple, repeatable, and reliable for even untrained users.

Based on anthropometric measurements of 100 human skeletons (195 total tibias) taken at the William Bass Skeletal Collection (University of Tennessee, Knoxville, TN), a safe zone for placement of an IO line with successful intravascular access was mapped for all the specimens (Fig. 6). The location of the safe zone of a typical patient based on measurements of all the specimens was identified as shown in the plot in Fig. 6.

Safe placement of an IO needle is ensured when it is inserted into the middle 50% of the width of the medial face, at a distance of at least 1.2 cm from the apex of the tibial tubercle, but less than 5 cm distal to the tubercle so as to avoid the nutrient foramen. Left and right tibia were considered as independent data points as they are separate extremities and develop independently of one another; doing so ensured that the device would be easily used on either side of a patient. Figure 7 shows a 3D model (Mimics, Materialise, Plymouth, MI) of the proximal human tibia with a 3D SolidWorks (Dassault Systèmes, Waltham, MA) model of the Tib-Finder incorporated. The models show how the average safe zone completely encompasses the opening in the Tib-Finder when it is placed on a patient's leg by a user.

Forty-eight users were tested on the intuition and skill associated with drilling a hole in the center of a circular target, perpendicular to an inclined surface. To eliminate variables associated with anatomy and delivery of IO Fluids, tests were performed on bone substitute foam with a triangular cross section. Figure 8 demonstrates graphically the alignment of the drill with respect to the plane and centering on the target area.

Figure 8(a) shows that the average displacement from on-center did not have a significant variance with respect to weighted skill score; this means that the skill with which a patient could drill the hole did not determine how well they could align the drill. The largest deviation from center was approximately 4 mm; this is less than 50% of the radius of the current opening in the Tib-Finder, so this user would still have been on-target for achieving intravascular access.

The ability of a user to align the drill, and subsequent hole, perpendicular to a surface is illustrated in Fig. 8(b). As seen in the figure, angular error averages less than 5 deg and is a maximum for one user at about 10 deg. It was determined from these results that the opening for EZ-IO needle should be at minimum 1 cm in diameter; there would still be enough distance from the center of the safe zone that a catheter placed near the edge of the opening would still penetrate the intramedullary canal and permit infusion of fluids. The feedback from human factors testing also led to addition of four crosshairs around the edges of the guide target area that would help visually align the drill tip.

Figure 9 depicts the current design-frozen Tib-Finder with pertinent features labeled. The semiflexible guide wings conform to the patient's leg and allow the provider to select between the left or right tibia. The tabs at the end of the guide wings allow proximal–distal orientation of the device using the patella. The combination of these guide structures and the IO guide's unique shape prevents misalignment due to rotation. The handle allows for grasping from any orientation enabling right and left handed users to operate the guide from either side, or at the foot, of the patient. A single hole indicates the insertion point for the IO drill (Fig. 9).

The Tib-Finder is an inexpensive, easily deployed, and useful guide for use with both manual and powered IO insertion tools. An alignment guide for placement of an intra-osseous infusion device has been developed, prototyped, and tested with positive results. The design is human centered and has been built around sound, diverse anthropometric data. The guide can be used with different sized drill bits provided with the EZ-IO system; further, the Tib-Finder is designed to interface with both powered and manual methods of IO line placement. To the best of our knowledge, this is a first-of-its-kind alignment tool, with the potential to allow for intra-osseous needle placement by anyone following a simple pictorial guide.

Briefly, the device iterations included replacement of the slot in Fig. 3(a) with a single hole to help reduce the risk that a needle would be placed in an inappropriate location. The shin guard concept was introduced that could flex to fit flush to the patient's leg when pressure was applied. Many possible options for the shin guard were explored, the first of which included a handle to allow for a robust, ergonomic grip. The handle gave the user control of the rotational alignment and provided a set point for applying pressure to hold the guide against the leg. Without a handle, the user had no set point on the guide with which to manipulate it and was also at risk for contacting the patient or accidentally sticking themselves with the IO needle. Each concept went through several iterations, performing basic testing with an EZ-IO gun to evaluate different variations on their basic geometry and different types of handles. The handle was chosen to be round to allow for grasping from any orientation and enabling right and left handed users to operate the guide from either side of the patient. The handle is made of a series of circular ribs to reduce material (Figs. 3(b) and 3(c)).

The addition of “wings” (Figs. 3(b) and 3(c)) on the end of the guide was made to orient it in relation to the patella. The wings sit on either side of the patella, mitigating possible misalignment due to rotation. Determining the correct length and geometry of the patellar wings was informed by the design team's experience with user testing of the device. The next modification came in the form of widening the patellar slots, shortening the wings, and flattening the guide slightly so as to require less flex to properly engage the tibia and subsequently less force on the part of the user to induce specified engagement. Users were observed in testing to be able to correctly place the IO needle, but the guide was easy to rotate away from the correct alignment. As a countermeasure to this newly discovered failure mode, the final iteration utilized widened patellar slots, but tightened up the bottom curve of the guide so it would still require some flex to sit flat on the patient's leg (Fig. 9). The results also suggested that human subjects could reliably drill a hole close to the center of the Tib-Finder's opening and with good enough alignment so as to be successful.

Future work involves deployment of the drill guide to users in prehospital setting, disaster scenarios, military situations, emergency rooms, and clinics near each investigator's businesses. This will ensure broad geographic distribution and breadth of patient body type, injury type, and ethnicity. The instructions and indications for use will be provided to all the users of the device, and it will be at their discretion to determine the appropriateness of using the Tib-Finder in a specific patient. Any further changes to the current model will be made based on feedback received from users who return an included feedback card.

The Tib-Finder is a one-time-use drill guide for use with IO catheters and is a nonsignificant risk class I device exempt from premarket notification (510 k) requirements; we are currently working to finish details related to the manufacture, sterilization, and packaging in accordance with FDA regulations. The Tib-Finder will hopefully be deployed in the coming months. It is the intent of the authors that this device be utilized to help those in the developing world get life-saving prehospital care otherwise not available without IV placement and also to improve the ease-of-placement as well as ensure accuracy of insertion for IO performed in the field and in hospitals.

The authors would like to acknowledge Dr. Saul Griffith and Otherlab, Inc., for their stewardship, guidance, and support during the initial phase of this project; the Office of the Dean of the Thayer School of Engineering and the Dartmouth College Technology Transfer Office have been generous in their support and guidance of our team; the White House Office of Science, Technology, Policy for inviting our team to the Ebola Summit in January, 2015; Christopher Magoon and Lindsay Holdcroft for their diligence and dedication for helping with the design of the Tib-Finder and handling logistics related to study planning and execution; and Matthew Crutchfield, Graphic Artist, Office of Medical Education, Marshall University School of Medicine.

• Defense Advanced Research Projects Agency (Contract No. HR0011-15-C-0028).

• A.H.S., S.D.R., and D.V.C. are cofounders of IOmetry, Inc.; the goal of this start-up is to manufacture and distribute, as well as train users in the operation of, the Tib-Finder.