Indications

IO infusion is indicated in situations requiring the rapid acquisition of intravenous (IV) access in which the establishment of conventional peripheral access is difficult or impossible. The situations in which it is most often used include cardiopulmonary arrest, shock, burns, and status epilepticus. In these situations, one or two attempts at standard peripheral access are usually made prior to placing an IO needle. In cardiopulmonary arrest, one should immediately gain vascular access via the IO procedure. In addition to its use in the hospital, IO access has been successful in the prehospital setting as well as in critical care transport.^4,5

The success rate with the technique is high, greater than 95% with experienced practitioners.⁶ Equal success using the new mechanical intraosseous devices has also been demonstrated.⁷ Most fluids and medications that can be given through a conventional IV line can be given via an IO infusion with comparable results. In the event of cardiac arrest or severe shock, IO access is at least as effective as peripheral venous access in providing fluids and medications to the central circulation. Studies have shown that commonly used resuscitation, antiepileptic, and antibiotic drugs all can be given effectively through an IO line. Three antibiotics produce subtherapeutic levels when given via an intraosseus line at standard IV doses: chloramphenicol, vancomycin, and tobramycin.⁸

In addition to the administration of fluids and medications, IO access can be used for certain clinical laboratory studies. No significant differences were found when comparing electrolytes, chemistries, pH, PCO₂, or hemoglobin from IO marrow specimen with either arterial or venous blood samples.⁹ A marrow specimen can be cultured in lieu of a blood culture.¹⁰ Finally, the marrow can be used for blood type and crossmatching.

Contraindications

IO infusion has few absolute contraindications. A fractured or previously punctured bone should not be used because infused fluid will extravasate and possibly cause compartment syndrome. Alternate sites in other bones can be used in such situations. Bone diseases such as osteogenesis imperfecta and osteopetrosis have been suggested as contraindications to intraosseous infusion,¹¹ but a case of successful IO access in a patient with osteogenesis imperfecta has been reported.⁶ Placing the needle into an area of cellulitis or burn could cause osteomyelitis or other infectious complications and is a relative contraindication although if limited sites are available, placing the IO needle through burned skin is acceptable.

Supplies and Equipment

Access to the bone marrow space is accomplished with one of several different types of needles. Conventional bone marrow needles (e.g., Jamshidi needle, CareFusion, San Diego, Calif.) work well. Needles made specifically for IO infusion use are available, including a straight needle or a needle with a threaded screw device (e.g., Sur-Fast, Cook Medical, Bloomington, Ind.). Usually a 15- or 18-gauge needle is chosen. The smaller 18-gauge should be used in infants. Studies have shown no significant differences in time required to insert the needle, success rate, or extravasation rates between standard and threaded IO needles.^12,13 If bone marrow or IO needles are not available, standard lumbar puncture needles can be used,^6,14 although they are prone to bend. In neonates, even a 19- or 21-gauge butterfly needle can be used.¹⁴ Needles with a stylet are preferred to prevent clogging of the needle by bone.

Two new mechanical devices appropriate for the pediatric population have been introduced: the Bone Injection Gun (B.I.G., Waismed Ltd., Herzliya, Israel) and the EZ-IO (Vidacare Corporation, San Antonio, Tex; see Figures 15-1 and 15-2). The B.I.G. is a spring-loaded device, while the EZ-IO is a small battery-powered drill; both of these penetrate the bone marrow more quickly as compared to the manual method.

Figure 15–1 Bone Injection Gun.

(Courtesy Persys Medial and Waismed USA.)

Figure 15–2 EZ-IO.

(Courtesy Vidacare Corporation.)

Other equipment required for IO placement includes a towel or sandbag, syringes with saline or heparinized saline flush solution, IV fluid and tubing, a T-connector or stopcock, and antiseptic prep solution (iodine). Optional supplies include a pressure bag and materials for local anesthesia (syringe with 25-gauge needle and 1% lidocaine).

Technique

The IO needle can be placed into the bone marrow at one of several sites, including the proximal tibia, distal femur, distal tibia, iliac crest, and sternum. The proximal tibia is the site most commonly chosen. The sternum has been used in adults but should be avoided in children because of the possibility of perforating the smaller chest cavity.¹⁵ In addition, placing the needle in the sternum can interfere with airway and circulatory resuscitative efforts. When the needle is placed in the proximal tibia, the insertion site is located by palpation on the flat anterior tibial surface 1 to 3 cm (two fingers breadth) distal to the tibial tuberosity (Figure 15-3). This site is chosen to avoid the proximal growth plate. The midshaft should not be used because of increased risk for fracture. Placing a towel or sandbag under the child’s leg helps stabilize the leg and makes insertion easier. The skin overlying the area should be prepped with antiseptic solution. Because IO lines are usually placed in obtunded patients, local anesthetic is not always necessary. Local anesthesia by infiltration with 1% lidocaine can be performed if the patient is awake. If a needle with a plastic sheath is being used, adjust the sheath so that an adequate length of needle protrudes beyond the sheath. Some authors have suggested inserting the needle at a 60- to 75-degree angle away from the tibial growth plate, but others recommend using a perpendicular or 90-degree angle. The perpendicular angle helps prevent the needle from sliding along the bone. The needle is advanced using firm pressure and a twisting or rotary motion until a “give” or loss of resistance is felt, indicating entry into the marrow space. The force needed to penetrate the bony cortex is considerable; the twisting motion helps significantly in needle insertion. One disadvantage of using a threaded needle is that the “give” or loss of resistance that occurs when the needle enters the marrow space may not be felt; however, the needle may be more secure in the bone once the needle is placed.²⁵ The stylet is removed and a syringe is attached to the needle to attempt to aspirate marrow. Correct placement of the needle should be confirmed to avoid extravasation. Aspiration of bloody fluid into the syringe confirms that the needle is correctly placed. Additional evidence of correct needle placement includes the observation that the needle stands upright in the bone without support and lack of resistance when flush solution is infused into the needle with the syringe. Sometimes, marrow cannot be aspirated even if the needle is correctly placed. If the “give” is felt on insertion and the needle stands alone in the bone but marrow cannot be aspirated, infusion of a small amount of fluid with the syringe can be attempted. Fluid should infuse easily with little pressure and without noticeable swelling of the soft tissues or extravasation of fluid.

Figure 15–3 Insertion of the intraosseous needle into the anterior tibia.

In small studies, placement of IO needles with the new mechanical devices have success rates equal to or higher than traditional manual methods, with the added benefit of easy use and less risk to the user.⁷ However, the high cost of this equipment may be a major limitation for use. In children, prehospital providers preferred the IO drill over the spring-loaded injection gun.^16,17

When using the spring-loaded injection gun, after choosing the desired needle depth of penetration by dialing to the patient’s age, the B.I.G. is positioned over the desired site with one hand while the other squeezes and pulls out the safety latch. The free hand is then used to activate the device at a 90-degree angle to the surface. The device is removed by pulling upward with a slight side movement to clear the needle. The stylet trocar is pulled out, then the safety latch should be slid over the needle and the apparatus taped as necessary for stabilization.

When using the IO drill, first ensure that driver and needle set are securely seated. After removing and discarding the needle set safety cap from the IO needle, position the driver at the insertion site with the needle set at a 90-degree angle to the bone and gently power or press the needle set until the needle set tip touches bone. Ensure that at least 5 mm of the catheter is visible. The bone cortex is penetrated by squeezing the driver’s trigger and applying gentle, steady downward pressure. The trigger should be released when a sudden “give” or “pop” is felt upon entry into the medullary space and a desired depth is obtained. If excessive force is used, the driver may stall and not penetrate the bone. Gentle pressure and the catheter tip rotation will provide the necessary penetrating action. After placement, remove the power driver and stylet and confirm catheter stability. A primed extension set is then attached to the catheter hub’s Luer lock and the apparatus is flushed. The manufacturer does not recommend attaching a syringe directly to the IO needle hub, to reduce the risk of needle dislocation.

The possible insertion sites are the same whether using the manual or mechanical needles. When the needle is placed in the distal femur, it should be placed approximately 2 to 3 cm proximal to the patella in the midline. In the distal tibia, the needle is placed 1 cm proximal to the medial malleolus in the midline, posterior to the saphenous vein. The proximal tibia is usually selected in children younger than 4 years, but the distal tibia has been recommended in older children in whom the proximal tibial cortex is thicker. The proximal humerus may be used for adolescents.

Maintenance

Once correct placement of the needle is confirmed, fluids and medication can be administered with a syringe via a stopcock or T-connector, or a standard IV infusion set can be connected to the needle. The needle is secured using gauze pads and tape. Some have suggested taping a clear plastic cup over the needle to help prevent dislodgment.⁶ The site should be observed carefully visually and by palpation for signs of extravasation, both immediately after placement and frequently (every 5 to 10 minutes) during use. If evidence of extravasation is seen, the needle should be removed to avoid compartment syndrome. If needle placement is attempted at one site and the cortex is penetrated but the line cannot be used because of extravasation, another bone must be chosen for subsequent attempts.

IO access is intended only for short-term use in emergency resuscitative situations; long-term use increases the risk of extravasation, compartment syndrome, and infection.^38,43 Therefore, once IO access is secured, efforts should be directed toward obtaining conventional IV access. Once alternate access is obtained, the IO needle should be removed, manual pressure applied for 5 minutes, and a dressing applied to the site.

Complications

Significant complications of IO infusion are rare. The most common complication is extravasation of fluid. The causes of extravasation include incomplete penetration of the bony cortex, movement of the needle such that the hole is larger than the needle, dislodgment of the needle, penetration of the posterior cortex, and leakage of fluid through another hole in the bone, such as a previous IO site or fracture. Extravasation of a small amount of fluid is usually not problematic, but compartment syndrome may result after extravasation.¹⁸ One report documented a compartment syndrome after infusion of only 35 mL of fluid via the IO line.¹⁹ Use of the IO line for prolonged periods and use of the IO line with pressure infusion appear to be risk factors for compartment syndrome. Compartment syndrome associated with IO infusion can result in the need for fasciotomy and amputation. Careful frequent observation of the IO site is necessary to detect extravasation and to prevent compartment syndrome. If extravasation occurs, the needle should be removed and the extremity observed for signs of compartment syndrome. Initial experience suggests the complications of the new mechanical devices are similar to the traditional IO needle.

Other rare complications include infection and bone fracture. Osteomyelitis, cellulitis, and sepsis have been reported in conjunction with IO infusion.²⁰ Risk for infection is increased when IO access is used in a bacteremic patient and when the line is used for a prolonged period. The risk of osteomyelitis is low. Although IO access is usually obtained in emergency situations, precautions to prevent infection should be taken, including use of sterile gloves and equipment and preparation of the skin with antiseptic solution (iodine or alcohol). Fracture of the bone has been reported.²¹ One report of fracture associated with IO infusion is the case of a 3-month-old in whom a 15-gauge needle was used; a smaller needle size (18-gauge) should be used in small infants.

There are some theoretical complications that are probably not significant in clinical practice. Fat embolism occurs with infusion into the bone marrow space. An animal study found fat emboli in the lung with IO infusion, but no changes were seen in oxygen saturation or intrapulmonary shunt.³⁵ There are no reports of clinical fat embolization associated with IO infusion in humans in the English literature. However, intravascular air was discovered on postmortem studies in two patients.²² Because of the risk of fat embolization, children with a right-to-left shunt may be at increased risk for cerebral emboli with IO infusion.⁶

The physical effects of infusion on the growing bone seem to be self-limited. Both animal studies and human follow-up studies of the bone and bone marrow show only short-term, minor changes and no long-term changes in bone structure, growth, or bone marrow.^23,24

Summary

IO infusion is a valuable means of obtaining temporary, emergency vascular access in the critically ill infant or child. It has a high success rate and is associated with rare complications. Complications can usually be prevented by using appropriate technique and vigilantly monitoring the site for extravasation.

Indications

Indications for an arterial catheter include the following:

Contraindications

Very few absolute contraindications for placement of an arterial catheter exist. The skin at the site of arterial access must be intact prior to insertion of a catheter. Any evidence of infection of the skin or underlying structures is a contraindication to catheter placement at this site. Other disruptions in skin integrity, such as burns, are a relative contraindication. Severe coagulopathy and systemic anticoagulation increase the risk of hemorrhage from unsuccessful arterial punctures associated with attempted arterial catheter placement, as well as the risks of bleeding at the site of arterial catheter insertion. These risks must be weighed against the potential benefits of improved monitoring when deciding to place an arterial catheter in a coagulopathic or anticoagulated patient.

A catheter should not be placed in an extremity with compromised perfusion. Evidence of adequate collateral circulation is desirable prior to placement of an arterial catheter. The traditional means of assessing collateral circulation to the hand is the Allen test. The radial and ulnar arteries are compressed until the distal extremity is blanched. Pressure over one artery then is released. Capillary refill should return to the distal extremity in less than 5 seconds. The test then is repeated by unblocking the other contributing artery after the distal extremity becomes blanched. Some practitioners question the utility of the Allen test because a normal test result does not guarantee adequate collateral circulation, nor does an abnormal test necessarily indicate possible complications.²⁵ The Allen test is considered less reliable for patients in shock. Additionally, an arterial catheter is often placed without assessing collateral circulation in emergent circumstances.

Procedure

Box 15-1 lists the supplies and equipment required for arterial catheterization.

Technique

The initial step in placing an arterial catheter is site selection. The radial, posterior tibial, and dorsalis pedis arteries are usually optimal sites, due both to the ease of access to these sites and to the typically good collateral circulation. Placement of the catheter in distal arteries of the extremities also allows for ease of site observation and hemorrhage control with direct pressure. Preductal placement in the right radial artery is preferred in infants with ductal-dependent heart lesions.

Catheters also can be placed in the axillary or femoral arteries if no peripheral sites are suitable. Insertion of a catheter into the axillary artery is technically more difficult than the other sites mentioned and is associated with a risk of brachial plexus injury due to hematoma compressing the neurovascular bundle.²⁶ Traditionally, many physicians have been reluctant to place arterial catheters for long-term use into the femoral artery, particularly in infants and young children, for fear of complications, most notably severe ischemia of the limb. A recent review of 745 femoral artery catheterizations in critically ill pediatric burn patients revealed a 1.1% rate of loss of distal pulse. In this retrospective study, limb ischemia was associated with younger age, smaller patient size, and increased severity of the burn injury. Patients who suffered limb ischemia were managed, for the most part, with immediate catheter removal and systemic heparinization. Three underwent thrombectomy. All patients in this series showed complete recovery of limb perfusion, except for one who required amputation of the fifth toe on the ipsilateral side.

Traditional teaching has held that the brachial arteries should not be used for arterial catheters because of the lack of collateral blood flow and risk of distal extremity ischemia. In a review of arterial catheter placements performed at a pediatric cardiac surgical center, the authors reported on the use of 386 brachial artery catheters in infants weighing 20 kg or less. None of these catheterizations were reportedly associated with permanent ischemic damage, and only three were associated with what was described as temporary loss of distal perfusion.²⁸ Despite these encouraging results, the complete lack of collateral circulation at the brachial artery requires careful consideration of risks and benefits before placement of a brachial artery catheter. Additionally, the superficial temporal arteries also should not be used because of poor collateral flow as well as the fact that retrograde flow from a catheter placed in this artery could result in showering of emboli into the cerebral circulation.

The selected site must be properly immobilized prior to placement of the indwelling catheter. If placing a radial artery catheter, the wrist is hyperextended 30 degrees to develop a straighter course and to bring the radial artery into a more superficial position. Typically, the radial pulse can best be palpated and catheter insertion should first be attempted in a position just proximal to the proximal crease, or usually about 1 to 2 cm from the wrist. Recently, the technique for placement of a radial arterial catheter has been expertly summarized in the “Videos in Clinical Medicine” series in the New England Journal of Medicine.²⁹ The following discussion of insertion technique can be applied to catheter placement at any of the sites discussed above. First, the site is prepared with a chlorhexidine solution and draped with sterile towels. Lidocaine (1% without epinephrine) is infiltrated locally. Alternatively, lidocaine/prilocaine (EMLA) cream can be used as a local anesthetic. Systemic narcotics and/or anxiolytics can be administered as well, although they should be used cautiously in a patient who is not being ventilated. Percutaneous placement of the catheter can be accomplished using one of several techniques. The first method uses a catheter over a needle, very similar to placement of a peripheral IV catheter. The needle is inserted through the skin at a 30-degree angle (bevel up or down). When a flashback of blood is obtained in the hub, the catheter is advanced another 1 to 2 mm. While holding the needle stable, the catheter is advanced over the needle into the lumen of the vessel. Blood should be flowing continuously into the catheter hub prior to attempting to advance the catheter. Once the catheter is inserted through the skin to the hub, a flushed Luer-Lok connector is attached to the hub while pressure is applied over the cannulated artery. Correct placement of the catheter is verified by aspirating arterial blood into a syringe attached to the connector. The catheter is then flushed and securely sutured or taped into position. A transparent dressing is placed over the catheter as a protective barrier. Before the final dressing is placed, many centers use a chlorhexidine-impregnated patch at the site of catheter insertion to decrease catheter-associated bloodstream infections.³⁰

Transfixation, the second percutaneous technique, involves inserting a catheter over a needle as described earlier. However, when a flashback of blood is seen in the hub, the needle and catheter are further advanced until the needle and catheter pierce the posterior wall of the artery and transfix this wall to the underlying structures. The needle is pulled out, leaving the catheter in place. The catheter is then slowly withdrawn until the tip is again intraluminal, with blood flowing back into the hub. The catheter is then advanced into the artery up to the hub. Catheter advancement can be facilitated by attaching a syringe filled with flush to a connecter and gently flushing as the catheter is advanced. The position of the catheter is confirmed, and the catheter secured as described earlier.

The final percutaneous method for catheter placement involves use of the Seldinger technique. A needle is used to pierce the anterior wall of the artery. When return of arterial blood is seen through the introducer needle, a guidewire is placed through the needle and advanced into the lumen of the artery. The needle is then removed, and a catheter is advanced over the guidewire into the lumen of the vessel. This method also can be used with a catheter over a needle. Confirmation of correct positioning and securing the catheter completes the procedure. A recent study in adults undergoing general anesthesia found no difference in the success rate or time to insertion when arterial catheters were placed using wire guides or without wire guides.³¹ Placement using the Seldinger technique is the preferred method when placing a catheter into larger vessels such as the femoral artery. When placing a catheter into a central artery using Seldinger technique, a longer catheter (e.g., 2.5 to 3 Fr, 5 to 8 cm) should be used. Such catheters are available in commercially made kits.

Recently, interest has grown in ultrasound-assisted placement of arterial catheters. A study of adult patients requiring arterial catheter placement in an emergency department setting demonstrated decreased time to insertion and decreased number of attempts at catheter placement using ultrasound-guided technique compared to traditional pulse palpation technique.³² A pilot study in pediatric patients undergoing surgery found that ultrasound-guided placement of radial arterial catheters had a higher success rate and required fewer attempts compared to traditional techniques.³³ However, a larger pediatric study that randomized patients receiving radial arterial catheters to placement via ultrasound-guided versus traditional techniques by pediatric subspecialty trainees or anesthesiologists found no difference in success rate at the first cannulation site, total number of attempts, or time to successful cannulation.³⁴

If attempts at percutaneous cannulation of an artery are unsuccessful, a cutdown approach is an alternative means of obtaining arterial access. A superficial incision is made perpendicular to the artery through the skin. The subcutaneous tissues are bluntly dissected parallel to the vessel using hemostats. When the artery is identified, the posterior wall is gently dissected away from the adjacent structures. Two loops are placed around the vessel: one proximal and one distal. These loops are used to elevate the artery during the cannulation process; they should never be used to tie off the vessel. After dissection is completed and the loops are in place, the artery is cannulated under direct visualization using the needle over the catheter method. The catheter is secured with a suture through the skin, and the wound is closed with interrupted stitches. If excessive bleeding persists, the proximal loop can have gentle traction applied in an attempt to control the hemorrhage.

Maintenance of an Arterial Catheter

To prolong patency of an arterial catheter, heparinized fluid is most commonly infused through the catheter. A common practice is to infuse 0.9% sodium chloride containing heparin 1 U/mL at 3 mL/hr, although slower infusion rates may occasionally be used in small infants with a need for fluid restriction. Studies of perioperative arterial catheters in adults have found no difference in the rate of catheter patency using heparinized or nonheparinized solutions,³⁵ but this does not take into account the smaller vessel size in children or the need for more prolonged catheter patency common to the care of the critically ill child. A meta-analysis of randomized controlled trials of heparin use to maintain either peripheral venous catheter or peripheral arterial catheter patency demonstrated a positive effect of heparinized fluid on duration of arterial catheter patency and the risk of clot formation.³⁶ Comparison of normal saline to 5% dextrose in water, both with heparin, revealed that infusion of saline resulted in longer catheter life.³⁷ Papaverine is a smooth-muscle relaxant that can decrease vasospasm when used in arterial catheter maintenance fluids. A randomized, controlled trial of the addition of papaverine (60 mg/500 mL) to routine arterial catheter fluids demonstrated a significantly lower rate of catheter failure and longer catheter life in the papaverine group.³⁸ This study evaluated patients from 3 weeks to 18 years of age and recommended avoiding the use of papaverine in neonates due to a perceived increased risk of intraventricular hemorrhage (IVH). More recently, the use of papaverine-containing arterial catheter fluid in neonates from 25 to 36 weeks’ gestational age was shown to prolong catheter life without an increased risk of IVH.³⁹ Despite these results, many institutions routinely avoid the use of papaverine in arterial catheter fluids in preterm neonates and in patients with traumatic brain injury or other preexisting intracranial hemorrhage for fear of causing or worsening an IVH.

The arterial catheter should always be visible so that any bleeding around the catheter site or caused by inadvertent disconnection of the tubing from the catheter, which can quickly result in significant hemorrhage, can be immediately observed. Securing the catheter with suture and using a Luer-Lok connector to connect the catheter to the tubing help to decrease the possibility of accidental detachment. The site of catheter insertion should be closely monitored for any signs of infection or compromised perfusion. Mottling of the skin proximal and/or distal to the catheter may be indicative of intraarterial thrombus formation, and discoloration of fingers or toes distal to a catheter may result from emboli. The catheter must be removed if any of these complications are observed.

For accurate direct measurement of blood pressure with an arterial catheter, one important consideration is the positioning of the transducer in relation to the catheter and the patient’s position. Standard practice for hemodynamic monitoring involves having the transducer zeroed to atmospheric pressure for accurate pressure measurement. The transducer is often maintained as close as possible to the level of the right atrium for accurate blood pressure measurements, particularly when the patient is in the supine position.⁴⁰ Studies in animal models have demonstrated that positioning the transducer level with the aortic root results in accurate measurement of mean arterial pressure regardless of patient position or catheter site, while placement of the transducer level with the catheter tip resulted in significant error in mean arterial pressure measurement when the patient was in positions such as reverse Trendelenburg.⁴¹ It is common practice at present to change the fluids and tubing every 72 hours. A study that investigated increasing the interval for changing arterial catheter fluids and tubing from 48 to 72 hours showed no increased risk for catheter-associated infection.⁴² The overlying dressing also is changed on a scheduled basis.

Inability to draw blood from a catheter or flattening of the waveform on the monitor is suggestive of either a kinked catheter or thrombus formation at the end of the catheter. If evidence of compromised perfusion distal to the catheter is present, the catheter should be removed. If no evidence of compromised perfusion is present, changing the catheter over a guidewire may be considered. However, strong consideration should be given to removing the existing catheter and placing a fresh arterial catheter in a new position, as changing a catheter over a guidewire has been associated with an increased risk of catheter-associated bloodstream infection (CABSI), at least in central venous catheters.⁴³

Complications

Complications related to arterial catheters include hemorrhage, thrombus formation, emboli, distal ischemia, and infection. Permanent ischemic complications related to radial artery catheters in adult patients are a very rare event.²⁵ A recent study used ICD-9 codes to define arterial catheter complications in a large multi-institutional administrative database in an effort to examine the prevalence of and risk factors for arterial catheter complications in critically ill children.⁴⁴ This study found a complication of some sort in 10.3% of patients with arterial catheters, most often infection/inflammation, and found a rate of thrombotic or embolic complications of 7.5%. Complications were more common in younger children and tended to be associated with longer hospital courses. Arterial catheter complications were independently associated with cardiac surgery, bone marrow transplantation, and dialysis, probably reflecting an effect of the severity of illness. In addition to bleeding, infection, and distal limb ischemia, an uncommon but well-recognized complication of arterial catheter placement is growth arrest due to physeal injury from extravasation, aneurysm formation, or ischemia.⁴⁵ Such injury can occur to the femur, tibia, radius, or ulna.

Catheter-related infections can be local or the focus for systemic sepsis. The risk of catheter-related infection has previously been thought to be lower for arterial catheters than for central venous catheters. Recent studies, however, have found very comparable rates of catheter-related infection in critically ill patients having both types of catheters.⁴⁶ The risk of arterial catheter infection has been related to the duration of catheter use and to placement of a catheter in the femoral artery. ^46–48 In one recent study of catheter-related infections in children, the presence of an arterial catheter was noted to be a risk factor for any type of catheter-related infection, but in this study the arterial catheter most likely represents a surrogate marker for greater severity of illness.⁴⁹ Regardless, however, the arterial catheter should be considered a possible source of sepsis, as is the central venous catheter, and strong consideration should be given to removing an arterial catheter when it is no longer absolutely necessary for optimal care.

Summary

Arterial catheters are a routine part of the monitoring of many critically ill children, and the ability to place an arterial catheter is a necessary skill for the pediatric intensivist. As with any procedure, the potential risks and benefits of arterial catheter placement should be carefully weighed prior to the procedure. Recent publications demonstrate that the rates of and risk factors for complications of arterial catheterization in critically ill children need further study.⁴⁴

Indications

Drainage of a pericardial effusion of any cause is absolutely indicated when cardiac tamponade is present. Often drainage is recommended if the effusion is large, even in the absence of tamponade, for diagnosis and fluid removal.⁵⁰ For small effusions, pericardiocentesis may be indicated for diagnosis alone. In pediatric patients, pericardial effusions most commonly occur with postviral or idiopathic pericarditis, but they are also seen with postpericardiotomy syndrome, collagen vascular disease, oncologic disease, and, rarely, uremia. Purulent pericarditis resulting from Staphylococcus aureus or Streptococcus pneumoniae infection can be seen in cases of concomitant pneumonia with empyema. Although rare in developed countries, tuberculous pericarditis can occur. Drainage of purulent pericarditis is indicated for relief of tamponade, prevention of constrictive pericarditis, diagnosis, and drainage of infection. With purulent pericarditis, open drainage may be more effective because of the difficulty in draining thick pus.⁵¹ If using a tube pericardiocentesis and drainage, instillation of a thrombolytic such as alteplase (recombinant tissue plasminogen factor) or urokinase may be considered with purulent effusions.⁵² Traumatic pericardial effusions secondary to penetrating trauma often require surgical drainage of the blood, because tamponade is common. Pneumopericardium secondary to pulmonary air leaks in mechanically ventilated patients is usually well tolerated hemodynamically but may require drainage, especially in small infants, because of the development of tamponade.

Contraindications

When acute tamponade is present, pericardiocentesis is unequivocally indicated. However, when elective or diagnostic pericardiocentesis is to be performed, the presence of a bleeding diathesis is considered a contraindication. The presence of aortic dissection is considered a major contraindication.⁵⁰ Lack of experience with the procedure is a relative contraindication for elective pericardiocentesis. Open drainage is preferred to closed drainage when the patient has traumatic tamponade and is in cardiac arrest.⁵³ When the effusion is loculated in a location not easily reached using the subxiphoid approach, needle pericardiocentesis is contraindicated because the risk of complications increases and the possibility of successful drainage becomes remote.

Procedure

Drainage of a pericardial effusion can be performed either by simple needle aspiration or by insertion of a drainage catheter. If the procedure is for diagnosis only and the effusion is small, then needle drainage is adequate. However, if tamponade is present, the effusion is sizable, or effusion likely will continue, insertion of a catheter for continuous drainage is indicated.

Equipment

Technique

Needle aspiration can be performed blindly in the event of a true emergency such as traumatic tamponade; however, the technique has higher complication and failure rates. Fluoroscopy has been useful but is cumbersome because it requires patient transport to a radiologic or catheterization suite in the absence of portable fluoroscopy.

Echocardiographic guidance is recommended for most pericardiocenteses because it can be done at the bedside and is logistically less complex. The most common technique uses transthoracic scanning, which can easily visualize the effusion. Echocardiographic scanning is indicated prior to needle drainage or catheter insertion for any reason except tamponade with cardiac arrest. Pericardial fluid can be identified by computed tomography or magnetic resonance imaging. Because these techniques are cumbersome and time consuming, they should not be used in acute tamponade because echocardiography is more readily available at the bedside and is usually fast in skilled hands. The echocardiogram can show the size of the effusion, its distribution around the heart including any loculations, the presence of fibrin or clots, and evidence of tamponade.⁵⁴ Tamponade can be diagnosed using two-dimensional imaging when the right atrium collapses during late diastole. Normally the right atrial free wall is concave throughout the cardiac cycle. With tamponade, the pericardial pressure exceeds the right atrial pressure at end-diastole and causes the right atrial free wall to collapse toward the center of the right atrium. The worse the tamponade, the longer into systole the collapse occurs.⁵⁵ Echocardiographic guidance in patients with tamponade has been shown in adults to have a 99% success rate in relieving the tamponade.⁵⁶ After the fluid has been echocardiographically evaluated, the patient is placed supine with the head elevated approximately 30 degrees. The subxiphoid approach is the safest and most common approach, although other approaches have been described. The approach is extrapleural and, in patients with normal anatomy, avoids major vessels such as the internal mammary, coronary, and pericardial arteries.⁵⁰ The subxiphoid and lower costal margin are prepared with 2% chlorhexidine. The area is draped in sterile fashion. Lidocaine local anesthesia is infiltrated at the junction of the xiphoid and the left costal margin. The needle is inserted at a 30- to 45-degree angle with the needle directed toward the left clavicle (Figure 15-4). The slip-tip syringe is attached and is aspirated continually while the needle is inserted (Figure 15-5). Needle advancement is halted when air or fluid is aspirated. If blood is obtained, analysis is necessary to determine whether the blood is of pericardial or intracardiac origin. Several techniques are helpful for this determination. The hematocrit of pericardial fluid will be lower than that of intracardiac blood, which will be equal to the patient’s hematocrit. Dropping a few milliliters of the fluid on gauze sponges determines whether the fluid will clot. Fluid that does not clot is pericardial; fluid that does clot is most likely intracardiac blood.⁵⁷ Another technique involves injection of small amounts of saline microbubble contrast (saline in a syringe that has been agitated) through the introducer needle while imaging with echocardiography.⁵⁸ If contrast bubbles are seen in the heart, then the tip of the needle is intracardiac. If bubbles appear in the pericardial sac, then the needle is appropriately placed in the pericardium.

Figure 15–4 Insertion of needle for pericardiocentesis at the junction of the xiphoid and the left costal margin, aiming toward the left shoulder.

(From Brundage SI, Scott BG, Karmy-Jones R et al: Pericardiocentesis and pericardial window. In Shoemaker WC, Velmahos GC, Demetriades D, editors: Procedures and monitoring for the critically ill, Philadelphia, 2002, Saunders Elsevier.)

Figure 15–5 In pericardiocentesis, the needle is inserted slowly, under continuous aspiration, toward the heart at a 15-degree angle to the skin.

(From Brundage SI, Scott BG, Karmy-Jones R et al: Pericardiocentesis and pericardial window. In Shoemaker WC, Velmahos GC, Demetriades D, editors: Procedures and monitoring for the critically ill, Philadelphia, 2002, Saunders Elsevier.)

During insertion the needle is guided using two-dimensional echocardiography. The echocardiographic probe is placed on the chest where the fluid is best seen. The needle tip is identified by ultrasound and followed as the needle is advanced.⁵⁹ Another technique involves mounting the needle on the echocardiographic probe, which has been placed in a sterile sleeve. The needle is advanced while the operator also handles the probe. This technique allows the use of locations other than the subxiphoid approach for insertion of the introducer needle, with the potential for better fluid visualization.⁶⁰

Previously, the most common guidance technique used electrocardiography to guide the needle. Alligator clamps were placed on the steel insertion needle and connected to an electrocardiogram monitor. The ECG complexes were visualized while the needle was inserted. If an injury pattern was noted in the ST segment, then the heart presumably had been touched by the needle. The needle was pulled back into the pericardial sac, and the ECG trace would return to normal. Although this technique is simple to perform, echocardiographic guidance offers more specific information about needle position and fluid location, allowing for a safer and more successful procedure.

Once the needle is determined to be well-positioned in the pericardial sac, if a catheter is to be inserted, then the J-wire can be passed through the needle as with standard Seldinger technique. The needle is removed, and the dilator is passed over the wire to open the tissues outside the pericardium and enlarge the puncture in the pericardium. The dilator is removed, taking care to leave the J-wire in good position. The catheter is passed over the wire into the pericardial sac. Its position can be confirmed echocardiographically. The wire is removed. The connecting tubing and three-way stopcock are connected and the fluid is aspirated. The tubing can be connected to a drainage bag for removal of fluid that continues to accumulate. A sterile sample of the aspirated fluid should be sent to the laboratory for appropriate analysis. The catheter is secured with suture and covered with an occlusive dressing. A chest radiogram should be taken at the end of the procedure, and daily to confirm the catheter position.

Maintenance

The catheter is simple to maintain. The dressing should be changed according to the ICU protocol for central venous lines. The fluid in the drainage bag should be measured and the amount recorded on a regular schedule. If fluid is no longer draining, then a small amount (1 to 2 mL) of heparinized saline can be infused into the pericardium through the stopcock after preparation with antiseptic solution. This process can release any fibrinous material occluding the catheter. If no fluid is forthcoming, then echocardiography can be performed to determine if residual pericardial fluid remains. If more fluid is present, flush the catheter again in an attempt to open up the catheter. If the fluid had originally been purulent, then instillation of a fibrinolytic agent may allow better drainage. If no fluid remains, the catheter can be removed, depending on the patient’s condition and the underlying cause of fluid development.

Complications

Cardiac puncture is not an uncommon complication of the procedure. If the ventricle is entered and the needle quickly withdrawn, then the injury is minor and of no clinical consequence. Coronary laceration occurs rarely, resulting in acute ischemia, and has been associated with death from the procedure.⁶¹ The pleural space can be entered with the insertion needle or the catheter, resulting in a hemothorax or pneumothorax. Hemoperitoneum and pneumoperitoneum can occur. Injury to the diaphragm, intestines, and stomach have been reported. Infection of the indwelling catheter can occur, but is rare because the catheter usually is not in place for more than 3 to 4 days. Cardiac arrest and death have been reported during pericardiocentesis.^{62, 63} In a retrospective study of pericardial drainage catheters in pediatric patients, Zahn et al.⁶⁴ noted only five complications in a total of 43 procedures. Four of the complications were considered minor and included myocardial perforation with no hemodynamic alteration, pneumopericardium, and ST-segment elevation. One critically ill neonate who had suffered a myocardial perforation during cardiac catheterization died. The results of the study indicated that pericardial catheter placement in pediatric patients is a relatively safe procedure. Most complications of pericardiocentesis with or without catheter placement can be prevented by careful guidance of the needle with echocardiography and greater experience of the practitioner.

Interpretation

Laboratory analysis of pericardial fluid includes cell count and differential, microbiologic evaluation, cytology, and blood chemistry workup, which may include protein, glucose, pH, lactate dehydrogenase (LDH) and triglycerides. Normally, pericardial fluid is clear and straw-colored. Cloudiness may indicate purulence resulting from infection, or chyle if the patient has undergone recent chest or cardiac surgery. Serosanguinous fluid can be seen with multiple diagnoses, including trauma, postpericardiotomy syndrome, collagen vascular disease, infection, and neoplastic disease.

The presence of white blood cells with a predominance of polymorphonuclear cells greater than 1000 per milliliter is indicative of bacterial infection. When the predominant cell type is lymphocytic, then tuberculous pericarditis should be suspected. In idiopathic or postviral pericarditis, the fluid contains fewer than 1000 cells per milliliter, and the cells are mostly monocytic or lymphocytic. Occasionally, the fluid is hemorrhagic with a preponderance of red blood cells. Neoplastic effusions may contain neoplastic cells, which may be missed on routine cell count but will be visible on cytologic examination. With purulent pericarditis, the protein level is high and the glucose level and pH are low. LDH is elevated with neoplastic effusions. With purulent effusions, the ratio of LDH in the fluid to LDH in the blood is increased to greater than 0.6. With chylous effusions, the fluid has an increased triglyceride level.

With acute nonpurulent pericarditis the fluid should be evaluated for viruses; Coxsackie viruses are the most common causes of viral pericarditis. Viral culture can be performed, although the yield is generally low. Polymerase chain reaction has been used successfully to diagnose viral and bacterial etiologies.⁶⁵ For purulent pericarditis, Gram stains and acid-fast bacilli stains should be performed, as should appropriate bacterial (both aerobic and anaerobic) and fungal cultures, as well as bacterial PCR analysis. If tuberculous pericarditis is suspected, adenosine deaminase (ADA) and PCR for mycobacteria are indicated.⁵⁰

Summary

Pericardiocentesis, with or without catheter placement, is a straightforward procedure that is indicated for relief of cardiac tamponade and for diagnosis of certain pericardial effusions. It is a lifesaving technique for patients with tamponade. Used in conjunction with guidance techniques such as echocardiography or electrocardiography, pericardiocentesis can be performed safely in patients of all ages.

Umbilical Artery Cannulation

Indications for umbilical artery cannulation (UAC) placement include the need for frequent monitoring of acid-base balance, continuous measurement of blood pressure, angiography, resuscitation, and exchange transfusions. Infusions of medications or fluids may be warranted when a UAC is present, but placement of a UAC is not indicated for infusions alone. UAC placement is contraindicated in patients with omphalitis, peritonitis, necrotizing enterocolitis, omphalocele, hypercoagulable states, or known ischemia of the perineum, or lower extremities.^66,67