Andreas A. Theodorou and Robert A. Berg
Rapid establishment of vascular access is necessary for aggressive fluid resuscitation and administration of medications such as catecholamines, antibiotics, narcotics, and sedatives during emergencies. However, attaining vascular access during a life-threatening illness in a child is difficult and often consumes precious time. An organized approach to vascular access can minimize this potentially life-threatening delay in treatment. This section discusses the priorities in vascular access during emergent, urgent, and stable situations. This chapter reviews various techniques for achieving vascular access and covers the relative indications and potential complications.
PRIORITIES OF VASCULAR ACCESS
Time is critical when attaining vascular access in life-threatening emergencies such as cardiopulmonary arrest or shock. Of course, any preexisting intravenous catheter should be utilized in the initial resuscitation efforts, regardless of how small such a catheter might be. If such access is not available during a life-threatening emergency, intraosseous access should be attained as rapidly as possible, especially in children under 6 years old. A practical approach is to pursue intraosseous and peripheral venous access simultaneously. However, time should not be wasted waiting for attempts at peripheral venous catheterization before attempting intraosseous access, because intraosseous access can be attained more rapidly and more reliably.1 Similarly, skilled clinicians may attempt placing central venous catheters during life-threatening emergencies, but such attempts should not preclude simultaneous attempts at intraosseous access. After attaining intraosseous access for initial fluid resuscitation and infusion of emergency medications, peripheral or central venous catheterization is the next priority in order to ensure a more reliable, long-lasting vascular access.
For urgent situations, such as fluid resuscitation of a child with compensated shock or dehydration, the risk-benefit ratio shifts. Generally, it is most appropriate to initially insert a peripheral venous over-the-needle catheter. If multiple attempts are unsuccessful, or if the child requires fluids or medications that cannot be given safely in a peripheral vein, central venous catheterization should be attempted by a qualified individual. Of course, if the child’s clinical condition deteriorates prior to achieving vascular access, priorities should be reassessed and it may be necessary to attain intraosseous access.
Relatively stable children may need vascular access for maintenance fluids or intravenous medications. Generally, peripheral venous cannulation with an over-the-needle catheter is adequate. If vascular access is necessary for more than 2 to 3 weeks, or if solutions to be infused can cause serious tissue injury if extravasated, a central venous catheter may be necessary. However, central venous catheterization entails added risks, and its inherent risks and benefits deserve consideration (Table 107-1).
Table 107-1. Vascular Access Techniques
PERCUTANEOUS PERIPHERAL VEIN CANNULATION
Small-caliber plastic catheters have allowed for increasingly easy and reliable peripheral venous cannulation. Small over-the-needle catheters (22–24 gauge) are available to cannulate even the small veins of the hand, foot, or scalp of neonates and premature babies. Such peripheral venous catheters are generally all that are necessary for short-term delivery of intravenous fluids or medications.2
The most common complication of peripheral venous cannulation is catheter displacement and infiltration of the tissues with the infusing fluid. Most intravenous solutions are isotonic or hypotonic and easily absorbed from the tissues. Infiltration with these solutions is generally treated by removing the catheter and elevating the limb. However, solutions that are very hypertonic (eg, those containing greater than 12.5% dextrose, 3% sodium chloride, or 8.4% sodium bicarbonate) or that contain irritating substances such as calcium or potassium salts, some antibiotics (eg, erythromycin), or medications with an extreme pH (eg, phenytoin) can cause substantial tissue injury leading to necrosis of the skin or subcutaneous tissues. Importantly, extravasation of vasoconstricting agents such as dopamine, epinephrine, and norepinephrine can result in profound local vasoconstriction and subsequent substantial tissue injury. Even without extravasation of the previously noted medications and fluids, local vascular injury can result in aseptic thrombophlebitis. This condition may serve as a nidus for suppurative thrombophlebitis. The risks of aseptic and suppurative thrombophlebitis increase with the duration of the indwelling catheter.
Selection of catheter size and peripheral venous site are important issues. For a patient in shock, the widest and shortest catheter is optimal, because longer, narrower catheters result in more resistance to flow.
For the trauma victim, it is preferable to insert the catheter into an uninjured limb or at least a limb apparently free from major vascular trauma. The greater saphenous vein, median cubital vein, and external jugular vein are often used, because they are relatively large and consistent in their anatomy. In older children, veins in the back of the hands and forearms are commonly used. Avoiding the dominant hand or the hand with the finger or thumb that the infant prefers to suck allows for improved patient comfort and function, although such choices are not always available.
INTRAOSSEOUS INFUSION
Although emergency intraosseous infusions were common in the 1940s and 1950s, the arrival of butterfly needles and plastic catheters and the widespread use of venisection led to virtual extinction of this technique for several decades. However, there has been a resurgence in the use of intraosseous infusions since the early 1980s.3 Its increasing popularity is primarily because intraosseous infusions can be performed rapidly and reliably. Medical personnel can usually attain intraosseous access in less than 1 minute during emergencies. The bone marrow cavity is effectively a noncollapsible vascular space, even in the setting of shock or cardiac arrest. Therefore, intraosseous access is the initial vascular access site of choice in patients with life-threatening problems such as cardiopulmonary arrest or de-compensated shock.1
Almost any medication that can be administered into a central or peripheral vein can be safely infused into the bone marrow, including crystalloid solutions, colloid solutions, blood products, and hypertonic solutions. In particular, all the medications the American Heart Association recommends for pediatric advanced life support can be safely and effectively administered via the intraosseous route, including catecholamine infusions. Pharmacokinetic variables, such as onset of action and plasma concentrations, are similar with intraosseous or peripheral venous administration in cases of circulatory shock or cardiac arrest with CPR. Emergency medications should be followed by a saline flush to ensure rapid delivery into the circulation. In addition, the initial bone marrow aspirate is a reliable specimen for venous pH and PCO2, blood typing and cross-matching, serum glucose, electrolytes, and blood cultures. Due to stasis in the bone marrow, the results of such studies may be less reliable after administering drugs through the intraosseous needle during CPR.
The greatest obstacle to successful intraosseous cannulation is psychological and originates from the natural reluctance by any inexperienced provider to force a needle into a child’s bone. Experienced providers may at times wrongly forfeit this quick and reliable alternative to undertake venous cannulation via rapid percutaneous central venous catheterization or peripheral vein cutdown, sometimes resulting in dangerous delays in starting therapy. Intraosseous cannulation is generally a safe procedure but can result in complications, including osteomyelitis, fractures, subcutaneous or intramuscular extravasation of toxic medications (eg, epinephrine, calcium), and compartment syndrome. Compartment syndrome is easily avoided by monitoring the insertion site for swelling and discontinuing the infusion if significant swelling occurs. Microvascular pulmonary fat and bone marrow emboli have been demonstrated but do not appear to be a clinically significant problem. The risk of such complications is acceptable in the dire circumstances of decompensated shock or cardiac arrest. Such risks, and the pain involved, may not be acceptable when the clinical indications are less stringent.
The most commonly utilized site for intraosseous access is the medial surface of the tibia, 1 to 3 cm below the tibial tuberosity. Alternative sites include the distal tibia above the medial malleolus, the distal femur, and the anterior superior iliac spine (Fig. 107-1). There are various styles of intraosseous needles. They are all designed to easily penetrate the bone cortex, and they all have a stylet to keep bone core from obstructing the needle during insertion. Aseptic technique and universal precautions should be followed. The needle should be twisted into, rather than pushed through, the bone marrow. Evidence for successful entrance into the marrow includes (1) the lack of resistance (or a “give”) after the needle passes through the cortex, (2) the ability of the needle to remain upright without support, (3) aspiration of the bone marrow into a syringe, and (4) free flow of the infusion without significant subcutaneous infiltration.1 Aspiration of bone marrow into the intraosseous needle is not always possible, especially in a very dehydrated patient. Infiltration of fluid into tissues around the bone is common when this route of infusion is used for a prolonged period of time or if the fluid is infused under great pressure. Infiltration will manifest as an enlarging leg circumference, fluid leaking out from the skin insertion site, or resistance to flow of the infusate.
CENTRAL VENOUS CATHETERIZATION
Central venous catheterization provides more reliable vascular access than peripheral venous catheterization. In addition, central venous catheters permit hemodynamic monitoring and laboratory sampling of central venous blood. However, the convenience of central venous catheters must be balanced against the added risks.
FIGURE 107-1. A: Preferred site for intraosseous cannulation in infants. B: Technique for insertion of the trocar. Note that the trocar is advanced with a twisting movement and in a caudal direction to avoid injuring the tibial growth plate.
Central venous cannulation is particularly suited for administering irritating or vasoconstrictive medications, which are diluted in the high blood flow of central veins, thereby limiting their contact with the vascular endothelium. Central venous cannulation may also offer a more expeditious alternative to peripheral venous access, particularly for patients who are in circulatory shock or cardiac arrest. Especially in these cases, a central venous catheter provides a means to monitor central venous pressure. In some circumstances, specialized central venous catheters may be used to monitor cardiac output, mixed venous oxygen saturation, pulmonary artery pressure, and pulmonary artery occlusion pressure (see Chapter 106).
In critically ill infants and children, polyurethane central venous catheters are generally inserted percutaneously through a large central vein such as the internal jugular, the subclavian, or the femoral vein. For chronic vascular access, particularly in patients requiring long-term chemotherapy, total parenteral nutrition, or antibiotics, Silastic catheters are often preferred, because they are less thrombogenic and have lower infection rates.2
Peripherally inserted central catheters (PICC) have gained enormous popularity, because they can be inserted easily at the bedside, usually through a peripheral vein (eg, brachial vein), and advanced into central location. The Broviac and Hickman catheters are placed in a central vein and tunneled out through a distant exit site in the skin. The Hickman catheters have Dacron cuffs that are very fibrogenic. The resultant fibrous scar around the Dacron allows for the catheter to be more securely anchored and decreases the risk of catheter infection from the skin. A last category of Silastic catheters are equipped with a proximal port that is embedded in the subcutaneous tissue (Port-A-Cath). These totally implantable venous access devices reduce the chance of microbial propagation through the catheter track and are most useful when only intermittent infusions are necessary.
There are unavoidable risks associated with all central venous catheters. During attempts to cannulate the internal jugular or subclavian veins, inadvertent needle puncture of the lungs can result in pneumothorax. Perforation of a large vessel with resultant leak into the pleural space can lead to hemothorax, and mechanical damage to the thoracic duct or thrombosis of the left brachiocephalic vein downstream from the insertion of the thoracic duct can lead to chylothorax. Central vein thrombosis and its attendant risks (venous congestion and edema, chylothorax) are particularly common in newborns and small infants because of the small size of their central vessels. For this reason, indwelling catheters have been progressively abandoned after cardiac surgery in favor of temporary atrial lines, which are placed at the time of surgery and removed during the immediate postoperative period once central access is no longer needed. In addition, injuries to nerves and arteries near the insertion site have been well documented.
When the catheter is open to the atmosphere and the patient creates negative intravascular pressure with a spontaneous breath, it is possible for air embolism to occur. This can be minimized by keeping the needle or catheter exit site in a dependent position (eg, Trendelenburg position during internal jugular or subclavian catheterization). The Seldinger technique, or guidewire technique, is commonly used to simplify catheter placement. A needle or trocar is used to puncture the vessel, followed by passage of a wire through the trocar. The trocar is then removed, and the catheter is then introduced over the wire. This technique is convenient but is associated with both an increased risk of vessel perforation by the guidewire, and catheter malposition because of the potential for the guidewire to take an unexpected course in the vascular system.
The most significant risk of central venous catheterization is infection. The risks of both thrombosis and suppurative thrombophlebitis increases with time and is especially high in children who suffer from hypercoagulability or immune deficiencies. The possibility of a hypercoagulable state should be considered in all critically ill children who develop catheter-related central venous thrombosis. Newer catheter materials (polyurethane, Silastic), heparin bonding, and low-dose heparin administration through the catheter may minimize the risks of thrombosis. Similarly, antibiotic bonding has been used to decrease the risk of infection.4 However, the most effective preventative measures are based on the conventional wisdom of placing central venous catheters only when indicated and removing them as soon as possible.
Local hemorrhage is common when removing the central venous catheters. Applying direct pressure as the catheter is removed can minimize the size of the hematoma. Air embolism is a rare, life-threatening complication associated with catheter removal. As during insertion, negative intrathoracic pressure with spontaneous respiration can result in negative intravascular pressure and entrapment of air in the vascular space. The major determinants of airflow are the difference in the atmospheric and vascular pressure and the size of the defect in the vascular wall. Therefore, the risk is greater with a large, strong adolescent than a small baby; and, similarly, the risk is greater with a large-bore catheter than it is with a small-bore central venous catheter. The risk of air embolism can be minimized by placing the patient in the Trendelenburg position, removing the catheter at the end of deep inspiration, and immediately covering the exit site with petroleum jelly gauze.
The most commonly used sites for central venous catheterization (see Figure 107-2) in acutely ill children are via the internal jugular (less frequently, the external jugular vein is cannulated). The subclavian veins are also often used by experienced clinicians, although the risk of pneumothorax or hemothorax is higher than with the other techniques. The femoral vein is the safest approach for less-experienced providers and, unlike in adults, does not seem to carry a higher risk of infection.4 Femoral cannulation offers clear anatomical landmarks, ease of hemostasia, and a safe distance from vital organs and sites of activity during resuscitation. Prior to catheterization of the internal or external jugular or subclavian veins, the child should be placed in the Trendelenburg position to help engorge the veins and minimize risk of embolism. Conversely, if the femoral vein is used, a slight reverse Trendelenburg position will engorge the veins and improve the success rate.
Several evidence-based steps can be taken to minimize the risk of infection.4 These include proper hand hygiene; preparing the skin with chlorhexidine; using a full-body sterile barrier, surgical masks, gowns, caps, and gloves for those performing the procedure; and removing unnecessary catheters. Using a checklist to ensure that these steps are followed has been associated with a marked decrease in the rate of catheter-related bloodstream infections.5,6
FIGURE 107-2. Three commonly used techniques for venous cannulation in children. A: Femoral vein. B: Internal jugular vein (the middle approach is shown here; the vein can be accessed via other alternative approaches). C: Subclavian vein.
Local anesthesia with 1% lidocaine should be provided subcutaneously unless the patient is anesthetized. A syringe and needle are used to locate the vein and aspirate blood. For femoral vein catheterization, the vein is located immediately medial to the artery, one to two finger’s breadths below the inguinal ligament. The introducer needle is inserted into the vein at a 45° angle. When blood is aspirated freely, the syringe is removed and the hub of the needle is covered with the thumb of the nondominant hand. A guidewire is then placed through the needle, and the needle is removed (Seldinger technique). A dilator is advanced over the wire to dilate the subcutaneous space and the entrance into the vein. The dilator is then removed, and the catheter is placed over the guidewire. After the catheter is introduced into the vein, the guidewire is removed, and blood is aspirated through the catheter to confirm intravascular placement. The electrocardiogram should be monitored during the procedure, because the guidewire can trigger arrhythmias when advanced into the heart. Proper position of the catheter should be confirmed. The method of confirmation depends somewhat on the intended site for the catheter tip. Blood-gas analysis can distinguish arterial from venous blood, but confusion may occur in patients with arterial hypoxemia. Transduction of vascular pressure during the procedure provides a more effective means of confirmation and is essential if the catheter is to be inserted into the pulmonary artery (see eFig. 106.1 and Chapter 106). A radiograph can also be used to confirm position, but sometimes a single projection is insufficient to determine the precise distal site of the catheter (eg, distinguishing right atrium from right ventricle).
ANALGESIA AND SEDATION FOR PROCEDURES
An important yet frequently neglected issue for any invasive procedure on a child is that of pain management and sedation. In the nonemergency situation, topical or local anesthetics should be used. The psychological milieu during the procedure should also be optimized. Music, toys, and interaction with family members can be important tools for improving comfort, thereby increasing the likelihood of cooperation and technical success. Moreover, minimizing discomfort during the procedure generally encourages better cooperation during the next potentially painful procedure for that child. However, when sedation is provided, personnel qualified to handle sedation-induced respiratory failure or shock should be present (see Chapter 113).