Journal of Clinical Anesthesia (2013) 25, 447–451

Original Contribution

Anesthesia through an intraosseous line using an 18-gaugeintravenous needle for emergency pediatric surgery☆,☆☆

Riyadh Khudeir Hamed MD (Senior Specialist Anesthesiologist & Specialist Pediatricand Neonatal Anesthesiologist)a,Sharon Hartmans MD (Resident in Internal Medicine)b,Marianne Gausche-Hill MD, FACEP, FAAP (Professor of Clinical Medicine;Vice Chair, Department of Emergency Medicine)c,d,⁎

aDepartment of Pediatric and Neonatal Anesthesia, Children’s Welfare Teaching Hospital, Medical City, Baghdad, IraqbDepartment of Medicine, Harbor-UCLA Medical Center, Torrance, CA 90509, USAcDepartment of Medicine, David Geffen School of Medicine at UCLA, University of California, Los Angeles, CA 90095, USAdDepartment of Emergency Medicine, Harbor-UCLA Medical Center, Torrance, CA; Los Angeles Biomedical ResearchInstitute at Harbor-UCLA, Torrance, CA 90509, USA

Received 29 July 2012; revised 15 February 2013; accepted 1 March 2013

B

0h

Keywords:Anesthesia: intraosseousroute;

Catheterization;Emergency surgery;Pediatrics

AbstractStudy Objective: To describe the success and complication rate of intraosseous (IO) access for deliveryof anesthesia with the use of an 18-gauge (G) intravenous (IV) needle.Design: Prospective study.Setting: Children’s Welfare Teaching Hospital, Baghdad, Iraq.Patients: 300 critically ill infants and toddlers, age 3 weeks to 16 months, requiring emergency surgeryfor intra-abdominal or pelvic conditions, in whom peripheral or central access was not obtainable.Patients presented for surgery between 2007 and 2010.Interventions: In 26 patients, the IO catheter was established when peripheral access was not obtainedat the outset of surgery; in 4 patients standard peripheral vascular access failed during the surgicalprocedure and IO access was obtained. An 18-G IV needle was placed into the proximal tibia andattached to an extension set with a 3-way stopcock to deliver anesthesia.Main Results: For 26 critically ill children and 4 other children, IV access failed during delivery ofanesthesia; vascular access was successfully obtained within minutes in all 30 infants (100%) using theintraosseous route. Ninety percent (27/30) of patients awoke immediately postoperatively in goodcondition; 10% (3/30) went to the pediatric intensive care unit (PICU) for further care due to theircritical preoperative condition. Complications associated with use of the IO route were considered minor(3/30 pts [10%]) and included extravasation of fluid in two cases and cellulitis in one.

☆ Supported by departmental funding only.☆☆ The authors have no conflicts of interest to disclose.⁎ Correspondence: Marianne Gausche-Hill, MD, FACEP, FAAP, Department of Emergency Medicine, Harbor-UCLAMedical Center, 1000W. Carson St.,

ox 21, Torrance, CA 90509, USA. Tel.: +1 310 222 3503, 222 6740; fax: +1 310 212 6101.E-mail addresses: drriyadhkh@yahoo.com (R.K. Hamed), mgausche@emedharbor.edu (M. Gausche-Hill).

952-8180/$ – see front matter © 2013 Elsevier Inc. All rights reserved.ttp://dx.doi.org/10.1016/j.jclinane.2013.03.013

448 R.K. Hamed et al.

Conclusion: The IO route provided for rapid delivery of anesthesia, induction, and maintenance in thisseries of critically ill infants undergoing emergency surgery when other vascular access routes failed.Few complications were noted. Intraosseous access was achieved through a simple technique using an18-gauge IV needle.© 2013 Elsevier Inc. All rights reserved.

1. Introduction

Securing a line for anesthesia in a hypovolemic patientprior to surgery may be very difficult – sometimesimpossible. When this happens, an intraosseous line (IO)is a safe, quick, and effective alternative [1]. The IO line isestablished in less than two minutes and is a significantlymore rapid site for venous access than central venousaccess [2].

Although the IO line is often considered as an alternativevascular access site in pediatric patients, it may be used in apatient of any age [3,4]. The IO route provides access to thesystemic venous circulation via the marrow or medullarycavity with its surrounding bone, and provides a non-collapsible entry point into the central venous circulation.Previous studies have demonstrated that the IO route is justas efficacious for medication administration as the intrave-nous (IV) route, with nearly identical pharmokinetics [5,6].The use of spinal needles and standard injection needles forIO access has been discouraged because of the flexibility ofthe needle wall (spinal) or possible obstruction of theinjection needle without a stylet by a bone plug [7,8]. Thecreation of specialized IO needles and new IO accessdevices, including devices that drill the IO needle into thebone (EZ-IO drill; Vidacare, San Antonio, TX, USA) andthe spring-loaded, impact-driven devices such as theBone Injection Gun (B.I.G.; WaisMed, Kansas City,MO, USA), have made placement of the IO line easy inthe emergency setting [2,8,9]. While there are numerousreports of IO line placement in various conditions in theprehospital and emergency department settings, there areconsiderably fewer reports of their use in the perioperativesetting [8].

We describe an approach to IO access in pediatric patientswho failed peripheral and central venous access attemptsduring anesthesia administration for emergency surgery in alarge teaching hospital in Baghdad, Iraq.

2. Materials and methods

This case series was submitted to the Institutional ReviewBoard (IRB) at Harbor-UCLA Medical Center for review; itwas determined that this historical cohort did not fall underthe category of research, as defined in 45CFR46.102(d), andis exempt. From 2007 to 2010, approximately 4,000pediatric patients presented to the Children’s Welfare

Teaching Hospital in Baghdad, Iraq, for emergency surgery.The hospital is a tertiary-care referral center with 200pediatric ward beds and 8 pediatric intensive care unit(PICU) beds. Most patients are transferred to this hospitalfrom neighboring villages and government health centers fora higher level of care. However, due to ongoing militaryaction in Iraq, patient transportation is often delayed forsecurity issues and patients present in a hypovolemic state,without peripheral venous access.

We describe a series of 30 pediatric patients whoreceived anesthesia via an IO line after several attempts togain peripheral or central IV access failed or a workingperipheral IV catheter failed during surgery. Since auto-mated IO drills were not available in Baghdad at the timeof this study, we used a standard 18-gauge (G) IV needlewith a handmade IV extension set. The gauge of the needlewas chosen to insure successful insertion without deforma-tion of the needle. A standard plastic cap from an IVcannula set was used to stabilize the physician’s thumb onthe end of the 18-G needle for insertion. The needle wasinserted only until a “pop” was felt and marrow waswithdrawn or fluids infused.

3. Results

Over a 3-year period (2008-2010), we placed IO lines in atotal of 30 pediatric patients. In 26 patients, we placed an IOline when peripheral or central IV access was not achievedprior to induction of anesthesia. In the remaining 4 patients,we placed IO lines after the start of anesthesia for a failed IVduring surgery. In all patients, IO access was a quick, easyand reliable way to obtain access for anesthesia for these life-saving operations.

The most common reason for emergency surgery inour patient population was intestinal obstruction(Table 1). Intraosseous placement occurred in 5 patientswith intussusception, 4 with congenital pyloric stenosisand severe dehydration, 3 with abdominal or hepatictumors, 3 cases of congenital diaphragmatic hernia, anda one month patient with vaginal bleeding due to arectovaginal fistula.

The most common site for IO access in children is theproximal tibia. The location was estimated by measuring onefingerbreadth below and medial to the tibial tuberosity alongthe proximal tibia [8,10]. The distal femur, and distal tibia arealterative sites for IO infusion in children, which we did not

Table 1 Type of emergency surgery performed on 26 patientsin whom anesthesia was initiated by intraosseous (IO) line ⁎

Type of surgical operation Cases(n)

Patients’age range

Exploratory laparatomy for intestinalobstruction

5 3 - 7 mos

Partial resection of colon & colostomyfor intestinal obstruction

5 3 wks - 9 mos

Reduction/resection of intussusception 5 4 wks - 4 mosPyloroplasty for pyloric stenosis 4 4 - 7 wksAbdominal & hepatic tumor resection 3 6 - 8 mosCongenital diaphragmatic hernia repair 3 3 - 6 wksRectovaginal fistula repair 1 16 mos

⁎ An additional 4 patients had an IO line placed after a peripheralline failed during surgery.

Fig. 2 Critically ill 7 month old infant with intestinal obstruction,with intraosseous line placed for induction of anesthesia.

449IO line in emergent pediatric surgery in Iraq

use. In all situations, the IO needle was placed successfullywithin a minute.

At the time of our study, our hospital lacked IO needlesor an automatic IO drill. An alternative technique wascreated using an ordinary 18-G IV needle. We attached theplastic cap cover from a disposable IV cannula set to thesyringe needle so that we could easily handle it duringpenetration through the bone cortex. After sterilization ofthe site, we advanced the needle through the bone cortexwith a twisting motion at a 90° angle until reaching themarrow (Fig. 1). After confirmation of proper positioning ofthe needle by aspiration of marrow, we secured the needlewith adhesive tape and gauze pads (Fig. 2). Then weconnected the needle to a handmade extension set with a 3-way stopcock for anesthesia infusion. In most cases, IOaccess was obtained in the awake state due to patients’critical condition and the risk of sedating a neonate prior to

Fig. 1 Critically ill newborn with intestinal obstruction, shownwith a proximal tibial intraosseous line placed for inductionof anesthesia.

securing vascular access. In the few cases where thepatient’s condition permitted, we used halothane to inducemild anesthesia prior to attempting to place an IO line.Perioperative antibiotics were used in all cases.

The induction and maintenance of anesthesia in thesesurgical operations was without serious complications. Forinduction therapy, we infused ketamine (1 - 2 mg/kg),followed by succinylcholine (1-2 mg/kg) and pancuroniumor atracurium. For maintenance therapy, we used halo-thane, and for reversal we used neostigmine (0.04 mg/kg)with atropine (0.02 mg/kg). We also administered bloodproducts, fluids, and hydrocortisone as needed through theIO line. After the procedure, we replaced IOs with IV linesas soon as possible, typically within a few hours ofsurgery. In all patients, IV or central lines were establishedwithin 6 hours of the completion of surgery. Ninetypercent (27/30) of our patients awoke immediatelypostoperatively in very good condition and were referredto the pediatric surgical ward. The other 10% (3/30) wentto the PICU and neonatal ICU for further care due to theircritical preoperative condition.

There were few complications in this series of patients. Intwo cases, we noted extravasation of fluid during surgery.When this occurred, we used the tibial bone of the oppositeleg and proceeded without further complications. One patientdeveloped cellulitis postoperatively.

4. Discussion

The idea that an IO line might be used to deliveranesthetic agents was first proposed in 1952 [11]. Tarrowet al suggested that in addition to fluids such as normalsaline, whole blood, and antibiotics, the IO route might be ofinterest to anesthesiologists due to the extensive blood

450 R.K. Hamed et al.

supply in the marrow. However, since their study, only alimited number of studies have reported the use of IOs forgeneral anesthesia in pediatric patients [12-16].

The largest study to date was reported by Neuhauset al and describes the semi-elective use of anestheticinfusion via an IO line [12]. A total of 14 children hadIOs placed when peripheral IV cannulation failed. In thismulti-center trial, authors inserted IOs into the proximaltibia of patients with either an automated drill or manualIO needle. They infused hypnotics, neuromuscular block-ing agents, opioids, cardiovascular drugs, antibiotics, andIV fluids without any significant complications.

Similarly, two case reports describe the use of IOs ininfants with cyanotic heart disease when peripheral IV accesswas unattainable. Steward and Kain report on the case of a3 month old infant who required ventriculoperitoneal shuntrevision [13]. The first attempt to place the shunt wascanceled after several experienced anesthesiologists wereunable to establish venous access. Two weeks later, when arevision was attempted, they placed an IO after attemptingperipheral IV placement for 20 minutes. Once the IO wassecured, they were able to successfully administer pancur-onium, lactated Ringer’s solution, and antibiotics withoutany complications.

Joseph and Tobias described the case of an 8 month oldinfant with cyanotic congenital heart disease undergoingdirect laryngoscopy and bronchoscopy to evaluate fortracheomalacia and vascular compression of the airway[14]. Three experienced pediatric anesthesiologists attemptedto secure a peripheral line. After 15 to 20 minutes ofunsuccessful attempts, they placed an IO and infusedpropofol for anesthesia, without any complications in thepostoperative period.

Alternatively, several reports have been published onthe use of IOs to induce anesthesia for emergency airwaymanagement in children. Katan et al described using anIO line to infuse lidocaine, thiopental sodium, andsuccinylcholine chloride for rapid-sequence intubation ina 6 month old child with elevated intracranial pressureand seizures secondary to suspected abusive head trauma[15]. Tobias and Nichols described two cases of IOanesthesia for rapid-sequence intubation [16-18]. Theyadministered atropine, lidocaine, succinylcholine, andthiopental via the IO route. In both cases, they notedmuscle relaxation in 45 seconds and were able success-fully to perform intubation.

Prior studies have demonstrated the effectiveness of IOsin children for use in elective surgery, emergency airwaymanagement, and in infants with cyanotic heart disease.However, this study demonstrated the effectiveness of IOsfor induction of general anesthesia in emergency surgery andadaptation of the 18-G IV needle for use in resource-poordeveloping countries.

In our patient population, we experienced few compli-cations, consistent with the published literature [1,8].Extravasation of fluid is a known complication and

potentially leads to tissue damage from medication orcompartment syndrome. In the two cases in which we notedextravasation of fluid, we quickly remedied the situation bychanging placement of the IO line to the alternate tibia.Osteomyelitis is another serious complication in IOcannulation; however, it occurs infrequently and we didnot encounter any cases within this study. In a metaanalysis,Rosetti et al observed that ostemyelitis occurred in only0.6% of more than 4,000 cases and that the overall failurerate of IO infusions was only 2.1% [1]. Finally, we did notencounter any complications with the use of an 18-Gneedle, such as obstruction by a bone plug. The flexibilityof the needle wall was not problematic either, as we used aneedle with a large enough gauge to prevent bending, andwe stabilized it with the plastic cap cover from a disposableIV cannula set.

If specialized IO needles or IO automatic drills are notavailable, IO access may be reliably achieved through asimple technique using an 18-G needle. Although it is not thefirst-line method for anesthesia induction, the IO routeshould be considered by pediatric anesthesiologists in caseswhere peripheral and central venous access have failed afterseveral unsuccessful attempts.

References

[1] Rosetti VA, Thompson BM, Miller J, Mateer JR, Aprahamian C.Intraosseous infusion: an alternative route of pediatric intravascularaccess. Ann Emerg Med 1985;14:885-8.

[2] Leidel BA, Kirchhoff C, Braunstein V, Bogner V, Biberthaler P, KanzKG. Comparison of two intraosseous access devices in adult patientsunder resuscitation in the emergency department: a prospective,randomized study. Resuscitation 2010;81:994-9.

[3] Foëx BA. Discovery of the intraosseous route for fluid administration.J Accid Emerg Med 2000;17:136-7.

[4] Leidel BA, Kirchhoff C, Bogner V, et al. Is the intraosseous accessroute fast and efficacious compared to conventional central venouscatherization in adult patients under resuscitation in the emergencydepartment? A prospective observational pilot study. Patient Saf Surg2009;3:24.

[5] Orlowski JP, Porembka DT, Gallagher JM, Lockrem JD, Van Lente F.Comparison study of intraosseous, central intravenous, and peripheralintravenous infusions of emergency drugs. Am J Dis Child 1990;144:112-7.

[6] Von Hoff DD, Kuhn JG, Burris HA 3rd, Miller LJ. Does intraosseousequal intravenous? A pharmacokinetic study. Am J Emerg Med2008;26:31-8.

[7] Engle WA. Intraosseous access for administration of medications inneonates. Clin Perinatol 2006;33:161-8.

[8] Tobias JD, Ross AK. Intraosseous infusions: a review for theanesthesiologist with a focus on pediatric use. Anesth Analg2010;110:391-401.

[9] Horton MA, Beamer C. Powered intraosseous insertion provides safeand effective vascular access for pediatric emergency patients. PediatrEmerg Care 2008;24:347-50.

[10] Kleinman ME, Chameides L, Schexnayder SM, et al. Part 14: pediatricadvanced life support: 2010 American Heart Association Guidelinesfor Cardiopulmonary Resuscitation and Emergency CardiovascularCare. Circulation 2010;122(18 Suppl 3):S876-908.

451IO line in emergent pediatric surgery in Iraq

[11] TarrowAB, Turkel H, ThompsonMS. Infusions via the bone marrow andbiopsy of the bone and bone marrow. Anesthesiology 1952;13:501-9.

[12] Neuhaus D, Weiss M, Engelhardt T, et al. Semi-elective intraosseousinfusion after failed intravenous access in pediatric anesthesia. PaediatrAnaesth 2010;20:168-71.

[13] Stewart FC, Kain ZN. Intraosseous Infusion: elective use in pediatricanesthesia. Anesth Analg 1992;75:626-9.

[14] Joseph G, Tobias JD. The use of intraosseous infusions in theoperating room. J Clin Anesth 2008;20:469-73.

[15] Katan BS, Olshaker JS, Dickerson SE. Intraosseous infusion of musclerelaxants. Am J Emerg Med 1988;6:353-4.

[16] Tobias JD, Nichols DG. Intraosseous succinylcholine for orotrachealintubation. Pediatr Emerg Care 1990;6:108-9.

[17] Selby IR, James MR. The intraosseous route for induction ofanaesthesia. Anaesthesia 1993;48:982-4.

[18] Waisman M, Roffman M, Bursztein S, Heifetz M. Intraosseousregional anesthesia as an alternative to intravenous regional anesthesia.J Trauma 1995;39:1153-6.

Reproduced with permission of the copyright owner. Further reproduction prohibited withoutpermission.