Chapter 45. Ultrasound-Guided Critical Care Procedures

The use of bedside ultrasonography has become an integral tool for evaluating and managing the critically ill patient. Improved image quality and increased portability of ultrasound machines have augmented the utility of physician-performed bedside ultrasonography in the emergency department and the intensive care unit. Not only is ultrasound useful diagnostically, but it can also shine a light into the darkness of invasive procedures traditionally performed with a landmark or “blind” technique. While initially the ultrasound-guided procedure may take more time as the practitioner's skills are developing, the ultimate benefits of fewer complications, less time to completion, and fewer attempts to complete the procedure make this technique worthwhile to master.1,2

Increased utilization of bedside sonography in the critical care arena has generated a large amount of research in the field. This has led to a paradigm shift in the way that procedures are performed at the bedside. Emblematic of this shift is the document released in 2001 by the Agency for Research and Health Care Quality, Making Health Care Safer: A Critical Analysis of Patient Safety Practices, which states that the body of evidence supports the use of ultrasound guidance for placement of central venous catheters.3

The performance of multiple critical care procedures may be improved with the addition of sonographic guidance. This chapter provides information about how ultrasound can be used to guide the following invasive procedures: central venous access, arterial line placement, pericardiocentesis, thoracentesis, paracentesis, lumbar puncture, endotracheal tube placement confirmation, and thoracostomy tube placement confirmation.

There are multiple transducers to choose from when performing a sonographic study. Selecting the correct probe can make the difference between a good image that is helpful and an image that is of poor quality and misleading. As a general rule, the higher the frequency, the better the resolution. The tradeoff is that high-frequency sound waves cannot penetrate deep into the tissues of the body. A high-frequency linear transducer is used to image superficial structures and is ideal for sonographic guidance when obtaining vascular access, performing lumbar puncture, and confirming placement of a thoracostomy or endotracheal tube. The curvilinear probe (2.5–5 MHz) is most often used for evaluating deep structures in the abdomen and pelvis. Because of its lower frequency, it results in good sound-wave penetration, but with a relative loss of resolution. The phased array probe (1–4 MHz) is ideal for intercostal imaging since it has a small footprint, a narrow superficial field of view, and a wider deep field of view (Figure 45-1).

When performing an ultrasound-guided procedure, the ultrasound machine should be positioned in such a way as to allow for easy visibility of the ultrasound screen by the practitioner. The ultrasound machine should be placed alongside the patient's bed directly in line with the operator's line of vision. This allows for minimal eye movement while looking at the ultrasound screen and performing the procedure. The machine should also be placed in a position so that there is enough cord slack to allow for adequate maneuvering of the transducer during the procedure (Figures 45-2 and 45-3).

When performing any invasive procedure, standard aseptic technique should be observed. Sterile gown, mask, cap, and gloves should be donned for all ultrasound-guided procedures. The addition of the ultrasound probe into the sterile field has the potential to compromise the sterility of the procedure. Therefore, transducer covers are essential to keeping the field sterile. There are many commercially available sterile covers that are particularly useful as the sterile cover usually extends to cover the cord as well. Alternatively, a sterile glove or occlusive dressing can also be used to cover the probe. However, care must be taken so the nonsterile cord will not contaminate the field. A nonsterile conducting medium (gel) can be placed between the probe and the sterile cover, but a sterile conducting medium such as sterile ultrasound gel, povidone iodine solution, or surgical lubricating jelly is placed between the probe cover and the skin (Figures 45-4 and 45-5).

Ultrasound guidance may be dynamic or static. With dynamic ultrasound guidance, the procedure is performed using continuous imaging. This is recommended for vascular access in order to ensure proper trajectory of the needle during the procedure. With static imaging, the anatomy is sonographically visualized and the needle entry point is marked on the skin. The transducer is then removed, the area is appropriately cleansed, and the procedure is performed in the traditional manner. The ultrasound probe does not need to be in a sterile sheath when using static imaging.

When first learning how to perform ultrasound-guided procedures, a two-person technique may be easier if dynamic ultrasound guidance is utilized. Using this approach, one person holds the transducer and guides the other person who performs the procedure. Once proficiency is obtained, a single-person technique can be employed, in which the transducer is held in the nondominant hand and the needle is held in the dominant hand.

Central venous access is frequently necessary to adequately manage the critically ill patient. Over 5 million central venous catheters are placed annually by physicians.4,5 Central venous catheters allow measurement of hemodynamic variables that cannot be measured accurately by noninvasive means and allow delivery of medications and nutritional support that cannot be given safely through peripheral intravenous catheters. Other indications for central venous access include lack of peripheral access and the need for aggressive fluid resuscitation. Many factors can make obtaining central venous access more difficult including body habitus, hypovolemia, congenital anomalies, poor vascular access due to history of intravenous drug use, and the presence of indwelling catheters. Typical sites for central venous access are the internal jugular vein, the supraclavicular and infraclavicular approaches to the subclavian vein, and the femoral vein.

Mechanical complications (e.g., arterial puncture, hematoma, pneumothorax, hemothorax) are reported to occur in 5–19% of patients, infectious complications (catheter colonization and associated blood stream infections) in 5–26%, and thrombotic complications (deep venous thrombosis) in 2–26%.6–9 Furthermore, coagulopathy, anatomic anomalies, anatomic deformity due to trauma, and operator inexperience can all contribute to a failure to cannulate a central vein. When compared with a traditional landmark approach, the use of ultrasound guidance has been shown to significantly reduce the mechanical complications.10–12 In a review of the literature, central venous catheter placement under ultrasound guidance was shown to decrease placement failure by 64%, complications by 78%, and multiple placement attempts by 40%.13 Achieving these lower rates of complication takes time and practice as placing an ultrasound-guided catheter requires developing additional hand–eye coordination.

As discussed above, appropriate positioning of the patient and the ultrasound machine is crucial for success. When cannulating the internal jugular or the subclavian vein, placing the patient in Trendelenburg will help to engorge the vein of interest and will allow for easier sonographic visualization and better blood return once the vein is punctured. For the femoral vein, reverse Trendelenburg will assist in this regard. One study described humming to be as effective as the Valsalva maneuver and Trendelenburg position for ultrasonographic visualization of the internal jugular vein and common femoral vein.14

Using the high-frequency linear array transducer, the area of interest should be prescanned to identify all pertinent structures as well as any structures in the field that should be avoided. In addition, confirmation that the desired vein is easily compressible demonstrates lack of an occult deep venous thrombosis. Recognizing the difference between the artery and the vein is extremely important in order to avoid arterial puncture. Veins are more easily compressible and have thinner walls than arteries. Doppler can be employed if there is doubt. Arteries will demonstrate characteristic arterial pulsations, whereas veins will demonstrate continuous flow.

For the short-axis approach, the depth of the center of the vessel from the skin surface should be measured. Once the depth is ascertained, the same distance is measured distally from the middle of the probe on the skin surface. This is where the needle should enter at a 45° angle. By triangulation, the needle tip should pierce the vessel. The needle is introduced into the ultrasound field under the center of the long face of the probe and is seen as a hyperechoic dot in cross-section. The true depth of the needle may be difficult to appreciate when using a transverse orientation. The transducer may need to be moved in a caudal or cephalad direction in order to find the needle tip. There are multiple visualization techniques that can be utilized to monitor needle placement and movement. Whether by direct visualization of the needle tip, seeing “ring-down” artifact, or noting tissue movement as the needle moves, one must be cognizant of needle location at all times, particularly when operating near other critical structures (Figures 45-6, 45-7, 45-8, 45-9).2,10

For the long-axis approach, the needle is introduced in the same plane as the long axis of the ultrasound transducer. The entire length and the depth of the needle are appreciated using this technique. However, the needle must be kept directly under the center of the transducer as there will be complete loss of visualization of the needle if either the needle or the transducer is moved out of plane.15 This is because the width of the beam generated by the transducer is very narrow. Recent evidence supports the use of the longitudinal approach over the transverse approach as it is associated with lower rates of posterior wall puncture rates when attempting to cannulate the internal jugular vein (Figures 45-10 and 45-11).16

Internal Jugular Vein

The internal jugular vein lies deep to the sternocleidomastoid muscle and is usually lateral and superficial to the carotid artery. Placing a catheter in this vein allows for central venous pressure monitoring and has the advantage of lower rates of pneumothorax when compared with attempting to cannulate the subclavian vein.5 In addition, catheters in the internal jugular vein have a lower infection rate than catheters in the femoral vein.17

To place a catheter in the internal jugular vein using sonographic guidance, place the patient in the Trendelenburg position. The ultrasound machine is positioned next to the patient's bed with the screen facing the head of the bed. With the probe, it is best to find an area where the carotid artery and the internal jugular vein are not in the same vertical plane. This will minimize the chance of arterial puncture via the posterior venous wall.11,13,18,19 Once the vessel is visualized and the surrounding anatomy appreciated, vessel cannulation proceeds as described above (Figure 45-12).

Subclavian Vein

When compared with other possible sites for central venous catheterization, the subclavian vein has the lowest rates of infection but also the highest rates of pneumothorax.5,20 For this procedure, the ultrasound machine should be positioned on the side of the bed opposite to the subclavian vein that will be cannulated. This will allow for easy visualization of the screen while placing the catheter. The subclavian vein crosses under the clavicle just medial to the midclavicular point. It is often difficult to visualize the subclavian vein at this location due to the strong echogenicity and posterior shadowing of the clavicle. To avoid this, one must visualize the vein more proximally or more distally with respect to the clavicle. It is possible to visualize the “venous lake” where the subclavian vein joins the internal jugular vein using the supraclavicular approach. It is also possible to visualize the subclavian vein inferior and lateral to the first rib where it is the proximal axillary vein (Figures 45-13 and 45-14). The vein may be cannulated in either of these locations. Due to the proximity of the subclavian vein to the lung pleura, a longitudinal approach is advised so that the entire needle can be visualized throughout the procedure.

Femoral Vein

Routine use of the femoral vein for central venous catheter placement should be avoided in adults due to higher rates of line sepsis and deep venous thrombosis formation when compared with other sites.5,7,21 However, the femoral vein is readily accessible in an emergent situation and can be useful as a compressible site in the setting of coagulopathy.5 To optimize positioning, the patient should be placed in reverse Trendelenburg with the leg externally rotated at the hip so as to increase the diameter of the femoral vein.22 The ultrasound machine should be placed at the side of the bed at shoulder level with the screen facing toward the patient's feet. Beginning the procedure in a transverse orientation just inferior to the inguinal ligament allows both the artery and vein to be visualized (Figure 45-15). The area where the greater saphenous vein drains into the common femoral vein usually provides the largest target. Usually, the vein will lie medial to the artery. In the femoral region, the vein and artery are adjacent to each other in the very proximal thigh, whereas more distally, the vein will be deep to the artery. Find the location where the artery and vein lie next to each other in order to avoid arterial puncture (Figure 45-16).23

Peripheral intravenous access is routinely performed in order to obtain blood for diagnostic tests as well as administration of fluids and medications. There are many factors that can make obtaining intravenous access challenging including obesity, intravenous drug use, and multiple previous peripheral access attempts.24,25 Ultrasound-guided peripheral vascular access is a safe and rapid method of obtaining vascular access and provides an alternative to either central venous access or multiple “blind” attempts.26 Furthermore, it has been shown that there is no increased risk of infection when performing ultrasound-guided peripheral access as compared with traditionally placed catheters.27

The high-frequency linear array transducer is utilized for obtaining vascular access of the superficial veins. Veins are round or oval anechoic structures that are easily collapsible with slight pressure of the probe on the skin. Color Doppler can be utilized to aid in differentiating an artery from a vein. The skin should be appropriately cleansed. The use of sterile ultrasound gel has not been shown to change infection rates.27 Once a suitable vein is identified, the static or dynamic technique employing either a transverse or longitudinal approach can be used to cannulate the vein. Frequently, by the time ultrasound is brought to the bedside, the patient has already undergone multiple attempts at vascular access. As a result, the external jugular vein and the veins in the proximal arm are often utilized to gain vascular access. If the brachial or the cephalic veins are cannulated, it may be necessary to use a longer (2.5 in) catheter as these veins may be beyond the reach of a standard intravenous catheter (Figures 45-17 and 45-18).28

Arterial catheter placement in a peripheral artery can be challenging, especially in a hypotensive patient with a weak or absent peripheral pulse. In the intensive care unit, the need for continuous blood pressure monitoring and frequent evaluation of arterial blood gas can make arterial catheters crucial to patient care. Traditionally, the arterial catheter is inserted via a technique using palpation of the pulse to guide the needle insertion. Difficulty with locating the artery or inserting the catheter subjects patients to multiple painful attempts. The use of ultrasound for arterial catheter insertion has been shown to increase first-pass success, thereby reducing time to insertion and the number of attempts.29

The linear high-frequency ultrasound transducer is used to identify the thick-walled pulsatile artery that is less compressible than the adjacent thin-walled and easily compressible veins. The addition of pulsed wave or color Doppler can be used to visualize the characteristic pulsatile arterial blood flow. Once the artery is identified, the catheter can be placed dynamically to monitor first the needle and then the guidewire entering the artery in real time. In cases where the radial artery cannot be cannulated, ultrasound guidance can also be used to cannulate the brachial, femoral, dorsalis pedis, or axillary artery.30,31

Cardiac tamponade is a life-threatening condition that can be remedied with pericardiocentesis. Traditionally, pericardial effusion and tamponade were diagnosed clinically using Beck's triad (hypotension, muffled heart sounds, and jugular venous distention), along with pulsus paradoxus, and a pericardial friction rub. However, many of these findings either occur late in the disease process or may be difficult to appreciate. Echocardiography has become the standard of care to diagnose a pericardial effusion and tamponade, and can also be used to localize the area with the largest fluid collection in anticipation of pericardiocentesis.32 Emergent pericardiocentesis should be performed when the patient sustains cardiac arrest or hemodynamic instability in the setting of a large pericardial effusion. Major complications of pericardiocentesis include cardiac chamber laceration, intercostal vessel injury, pneumothorax, sustained ventricular tachycardia, and death.33,34 Ultrasound guidance can make this dangerous, yet life-saving procedure, significantly safer.33

The patient should first be imaged using standard cardiac windows with the phased array transducer. This also allows for evaluation of global cardiac function. Cardiac tamponade can be diagnosed on ultrasound by observing right ventricular collapse during early diastole or invagination of the free wall of the right atrium during end diastole. A plethoric inferior vena cava with little change in diameter during respiration as well as alterations in transvalvular flow velocities may also be visualized (Figure 45-19).35

The subxiphoid and the parasternal approaches are most commonly used for pericardiocentesis The decision to use one rather than the other lies mainly with operator experience and where the largest amount of fluid is. The subxiphoid approach is most often used with the static technique, while the parasternal approach is more commonly performed as a dynamic ultrasound-guided procedure (Figure 45-20).

For a parasternal approach, the patient can be supine, with the upper body elevated 30–45°, or the patient can be placed in the left lateral decubitus position. The patient should be prepped and draped using standard sterile technique. The ultrasound transducer can then be placed in the parasternal long-axis orientation at the third or fourth intercostal space just to the left of the sternum to visualize the largest pocket of fluid that is usually between the probe and the anterior wall of the heart. The distance from the skin to the pericardial space should be measured to judge how far the needle will be advanced in order to enter the pericardial space.36 Care must be taken to avoid the left internal mammary artery that lies 3–5 cm lateral to the sternal border. Once the ideal entry path has been determined, local anesthesia should be infiltrated at the entry site and along the proposed path for the needle. The longitudinal approach is recommended for this procedure so that the needle can be visualized throughout the procedure. The needle should be large bore, 18 gauge or greater, and preferably sheathed with a Teflon catheter to allow for continuous drainage of the effusion. A thoracentesis tray usually contains this type of catheter. As the needle is inserted, the tip should be followed carefully with continuous sonographic visualization. Once a “flash” of fluid is obtained, the needle can be advanced a few millimeters and the fluid can then be aspirated with the needle or the Teflon sheath can be advanced while the needle is removed for continuous drainage.37

When employing a subxiphoid approach, either static or dynamic sonographic guidance can be used. For the static technique, the probe is positioned in the subxiphoid region. The effusion should be visualized with the largest pocket centered on the screen. The entry point, direction of the needle, and depth necessary to reach the fluid should be noted. After the probe is removed, the needle can be introduced as described above.36,38

Pleural effusion is a relatively common entity among critically ill patients and thoracentesis is the most common interventional thoracic procedure.39 The etiology of the collection can be elusive and pathologic examination of a fluid sample is often diagnostic. Potential complications of thoracentesis include pain, pneumothorax, vasovagal reactions, reexpansion pulmonary edema, inadvertent liver or splenic laceration, as well as infection and hemothorax.39,40 Ultrasound can be used not only to detect an effusion but also to guide drainage, which may be necessary to perform emergently if the pleural fluid is causing respiratory distress. Ultrasound is more sensitive than chest radiography for detecting pleural effusions. It has also been shown to increase the success rate of thoracentesis and to decrease the rate of complications,39,40 including pneumothorax and lacerations of the liver and spleen.41

To detect a pleural effusion by ultrasound, place the patient in a supine position with the head of the bed at 45°. The low-frequency curved abdominal probe should be placed longitudinally in the posterior axillary line just inferior to the nipple so that the curved hyperechoic line of the diaphragm above the liver or spleen can be visualized. A pleural effusion will be represented by an anechoic collection of fluid superior to the diaphragm. When no effusion is present, the spinal stripe will terminate at the diaphragm as the ultrasound waves are refracted by the air-filled lungs. However, when an effusion is present, the fluid collection allows for visualization of structures normally obscured by the lungs, namely, the thoracic spine superior to the diaphragm. Visualizing the hyperechoic spinal stripe superior to the diaphragm indicates that fluid is present in the chest, as does an anechoic collection with sharp angles superior to the diaphragm (Figure 45-21). Once an effusion has been detected, a more thorough evaluation of the chest should be performed. The movement of the lung with the respiratory cycle and the location of the diaphragm and abdominal organs should be noted so as to avoid these structures when a thoracentesis is performed.

With a cooperative awake patient, the patient can be placed in the sitting position at the edge of the bed with his or her arms folded on a tray table. The posterior hemithorax should be scanned with a small footprint phased array or a microconvex probe from the inferior scapular border to the upper lumbar region, and then from the paraspinal region to the posterior axillary line in order to delineate the extent of the fluid collection and the area of the largest pocket.41 A hyperechoic line, representing the pleura, should be visible between the ribs. Noting the depth of the pleura will be useful to determine the length the needle will need to traverse to successfully aspirate fluid.35 After the effusion is mapped out, the critical structures of the diaphragm and the lung are located, and the entry point is determined and marked. A static approach is employed and the procedure proceeds in the normal fashion. Care must be taken to ensure that the needle is placed superior to the diaphragm. The hyperechoic diaphragm should be visualized and the needle should be introduced at least two rib spaces above this structure to ensure that the peritoneal cavity is not violated. The procedure can also be performed dynamically, visualizing the needle as it enters the pleural space. If a thoracostomy tube is to remain in the thoracic cavity, a location in the midaxillary line will be more comfortable for the patient when lying in the supine position.

A sedated or intubated patient requiring a thoracentesis should be placed in the supine position with the arm abducted and the stretcher in the reverse Trendelenburg position. The puncture site will be in the lateral chest in the midaxillary line similar to the location commonly used for chest tube placement. Ultrasound can help to target the largest fluid pocket and to avoid critical structures. If the patient is mechanically ventilated, temporarily decreasing the tidal volume during the procedure may reduce the incidence of pneumothorax. While pneumothorax is a significant complication in the mechanically ventilated patient, the incidence is decreased with use of ultrasound.43,44

Paracentesis is performed in the critically ill patient with ascites for diagnostic or therapeutic reasons, or both. US guidance has been shown to confer a much higher success rate when compared with a landmark-based technique (95% vs. 65%).35 An additional benefit of ultrasound guidance is visualizing little or no free fluid in the abdomen in a patient with a markedly distended abdomen that was thought to be due to ascites. This discovery can spare the patient an invasive procedure.

With the patient supine or in a slightly left lateral oblique position and the head of the bed slightly raised, the lower abdomen should be scanned in two orthogonal planes with the low-frequency curved transducer. The left lower quadrant has been traditionally preferred as a gas-filled cecum or an appendectomy scar inhibiting the free flow of fluid may be present in the right lower quadrant, although ultrasound guidance allows for identification and avoidance of these structures so either lower quadrant may be utilized. The largest fluid collection site should be noted and the entry point marked. A fluid collection at least 3 cm in diameter is considered adequate for drainage. Static guidance is most commonly used for this procedure, although one may use dynamic guidance to observe the needle entering the peritoneum (Figure 45-22).

Structures to note and avoid when scanning the abdomen are the inferior epigastric artery in the abdominal wall, the bladder, and the intestines. The thickness of the abdominal wall and the depth of the fluid collection should be noted as well. Also note that a large loop of fluid-filled bowel should not be mistaken for a fluid collection. Fluid-filled bowel has a hyperechoic wall surrounding it and exhibits peristalsis, whereas peritoneal fluid will be located outside the intestinal walls.

Unrecognized esophageal intubation is relatively infrequent, although in those rare occurrences, morbidity and mortality are significantly increased.44 There are multiple methods employed at the bedside to confirm endotracheal intubation including direct visualization of the tube passing between the vocal cords, chest rise after intubation, auscultation over both lungs, and end-tidal carbon dioxide monitoring. However, any single method is not entirely reliable. Confirmatory methods may be inaccurate, especially in patients suffering from cardiac arrest or traumatic injury or those who have recently vomited or bled.44,45

Ultrasound has recently been shown to be useful as a bedside adjunct for confirming endotracheal intubation. Using the high-frequency linear array transducer, ultrasonography should be performed dynamically during the intubation procedure. Just prior to intubation, the vocal cords can be visualized as a triangular structure. The probe is held transversely over the cricothyroid membrane. As the endotracheal tube is passed, a brief “snow storm” should be seen, after which the transcricothyroid membrane should appear round with posterior hypoechoic shadowing.44,46 If the tube has instead passed through the esophagus, the unintubated trachea produces a reverberating type of artifact knows as “ring down”. Usually, the esophagus is not visualized sonographically, as it is collapsed. However, in the setting of an esophageal intubation, the esophagus may be noted to be distended by the endotracheal tube. In addition, confirmation of bilateral lung sliding should be performed by placing the linear transducer longitudinally over both hemithoraces in order to check for pleural sliding. The presence of bilateral sliding provides confirmation of an endotracheal intubation and helps to differentiate a right mainstem intubation from a main tracheal intubation. The bilateral absence of lung sliding may indicate an esophageal intubation (Figure 45-23 and 45-24).44,46–49

Subcutaneous placement of the chest tube is a known complication of tube thoracostomy. While chest radiography is routinely used, differentiating tubes that are extrathoracic but appear intrathoracic in the anteroposterior radiographic view can be challenging. Computed tomography of the chest is more sensitive and specific in identifying correct tube placement. However, this imaging modality requires a patient to be moved to the radiology suite, which may be inadvisable in an unstable patient. Ultrasonography has recently emerged as a sensitive and specific modality for verifying intrathoracic versus extrathoracic tube thoracostomy placement.

Using a linear array high-frequency transducer placed in a transverse orientation relative to the chest tube, the tube will appear as a hyperechoic arc. Starting at the site where the tube pierced the skin, the tube is followed and the disappearance of the arc indicates that the tube entered the pleural space. Visualization of the arc superior to the thoracostomy site for the length of the tube indicates that the tube is in the subcutaneous tissue.50

Lumbar puncture is a commonly performed procedure used to aid in the diagnosis of meningitis, subarachnoid hemorrhage, and other neurologic emergencies. Traditionally, the procedure relies on identification of bony landmarks. However, these landmarks may be difficult to palpate or identify due to body habitus, contractions, or inability to properly position the patient. Furthermore, using Tuffier's line between the iliac crests to identify safe lumbar interspaces may be inaccurate.51 Ultrasound guidance has been shown to increase the success of lumbar puncture in patients with elevated body mass index52 and may be helpful after multiple failed attempts in the nonobese patient.

To perform the procedure, the patient is placed in the lateral decubitus or the sitting position. The linear high-frequency probe is preferred. In an obese patient, a low-frequency probe may be used if deeper sonographic penetration is necessary to visualize the bones. The probe is placed at the top of the gluteal fold in a longitudinal orientation and the hyperechoic bony sacrum should be visualized. Moving superiorly, the first break in the hyperechoic line is the L5–S1 disc space. The L5 spinous process appears as a hyperechoic convex line. Moving the transducer superiorly, the three lower spinous processes and the intervening disc spaces can be mapped out. The probe can be moved laterally to find the lateral margins of the spinous processes as they disappear from view. Also, the probe can be turned 90° to better appreciate the lateral margins of the spinous processes (Figure 45-25).53,54

With indelible ink, the central line of the spine and each spinous process can be drawn on the skin. These marks can be used to proceed with the lumbar puncture in the traditional manner using the pen marks instead of the bony landmarks as a guide.

There is a growing body of literature demonstrating the value of ultrasonography in the care of the critically ill patient. Ultrasound is widely available, portable, repeatable, relatively inexpensive, pain-free, and safe. It should be considered in conjunction with traditional methods when performing invasive procedures.