Ultrasound-guided peripheral venous cannulation in critically ill patients: a practical guideline

Background Up to one-third of critically ill patients have difficult intravenous access (DIVA). This occurs often in obese patients, those with generalized edemas or in patients with previous venous cannulations. In DIVA patients, the conventional technique often fails. In contrast, ultrasound-guided cannulation has demonstrated a high success rate, improving patient satisfaction and even a reduction in the need of central venous lines. However, a high rate of premature catheter failure has been shown in cannulations performed by ultrasound guidance and thus a comprehensive knowledge of several aspects related to this procedure is mandatory to improve cannulation success, avoid complications and lengthen the survival of the catheter. Main text Several practical issues related to peripheral venous cannulation are described: peripheral venous anatomy, vein size and catheter selection, distance from skin to vein, insertion angle and selection of the catheter length, cannulation technique itself (out-of-plane or in-plane) and checking catheter position. Conclusion Key concepts regarding ultrasound-guided peripheral vein cannulation should be well known for practitioners, aiding in improving cannulation success and catheter dwell time, and avoiding complications.


Background
Peripheral venous cannulation is essential to provide care for the patients in the emergency department or critical care unit. While most intravenous catheters are placed using the conventional technique (i.e., seeing and/or palpating the vein), up to one-third of the patients have difficult intravenous access (DIVA) [1]. This group often involves patients with generalized edemas, obese, those with multiple previous cannulations or intravenous drug users [2][3][4][5]. For patients who have DIVA, ultrasound (US)-guided cannulation has shown an overall success rate higher than 90%, compared to 25-30% using the conventional technique [2][3][4]6], and also aids in reducing the need for central venous lines [4,5]. Patient satisfaction also improves using US guidance [2]. In spite of that, the rate of premature catheter failure (PCF), which may account for as high as 50% within 24 h of catheter placement, is higher with US-guided cannulations (45-56%) compared to the conventional technique (19-25%) [1,6,7]. Infiltration is the leading cause of catheter failure; catheter dislodgement and thrombophlebitis are also common [1,6]. Key concepts regarding veins, catheters and the technique itself should be considered by practitioners to improve success, reduce complications and improve dwell times in US-guided peripheral intravenous cannulation, and these are provided in this article, and are summarized in Table 1 as well. Basic general knowledge for US-guided vascular cannulation is shown in Fig. 1. to the skin and travel without an accompanying artery or nerve. In contrast, deep veins (which may be paired, as seen in brachial veins) are found at the neurovascular bundle, and thus are accompanied by an artery and a nerve ( Fig. 2 and Additional file 1: Video S1). From these anatomical points, some key concepts advocate for the use of superficial veins instead of deep veins.
Using superficial veins provides a short pathway to cannulation, leads to dwell a higher proportion of the catheter inside the vein (an issue intimately related to PCF, see below) and as a safe issue, avoids needle-stick injury of the artery or nerve. Cannulation of deep veins is also associated with a greater risk of catheter dislodgment when compared to cannulation of superficial

Key concept 2: Select patent veins
A fully patent vein is sine qua non for cannulation. This is demonstrated by applying slight compression forces over the skin with the transducer and observing veins that collapse easily (Additional file 2: Video S2). In contrast, a thrombosed vein is partially or totally non-compressible, is filled with thrombotic material (Additional file 3: Video S3) and thus is discarded for cannulation. After applying a tourniquet, a stagnant blood flow may be observed within the vein lumen in two-dimensional imaging, and this should not be confused with a thrombus (Additional file 4: Video S4). Distal compression aids in clearing this stagnant blood from the vein and ruling out a thrombus when vein patency is in doubt. Since peripheral veins have low blood flow velocity, spontaneous signal may not be observed in color Doppler. In these cases, distal compression allows to squeeze the blood from the vein, elevate blood flow velocity, thus aiding in the demonstration of flow in patent veins.

Key concept 3: Determine vein size-catheter size
An optimal vein size is required to improve cannulation, and the size recommended in the literature is at least 4 mm in anteroposterior (AP) diameter [9] (Fig. 4). This suggested vein diameter, although important, should not be used in practice in a strict manner, since smaller veins can still be cannulated successfully, as seen in some studies [4,6]. The vein size fulfills not only an established role in cannulation success (i.e., large veins are easily  observed, as well as the needle within the vessel), but also aids in guiding catheter selection. As a rule of thumb, AP diameter indicates the upper limit of the external diameter of the catheter which can be used, considering that up to one-third of the vein lumen should be occupied by the catheter [10]. Thus, for example, a 4-Fr catheter (with an external diameter of 1.3 mm) is the maximum size for a 4-mm vein.

Key concept 4: Determine vein depth-insertion anglecatheter length
As mentioned before, apart from avoiding the injury of the artery or nerve, practitioners should select superficial veins to guarantee a short pathway to cannulate the vein. The maximum suggested distance from skin to vein is < 16 mm [9,11], while < 12 mm can be considered ideal [1] (Fig. 5). This distance supposes a 90° needle insertion related to the vessel, and thus, the real or "corrected" distance of the needle traveling to reach the vein can be approached performing Pythagorean assumptions, which are entirely true for a needle insertion angle of 45°. This distance is equal to 1.4 multiplied by the vertical distance (Fig. 6). For example, a vertical distance of 12 mm equals 16.8 mm using a 45° insertion. However, in practice, this length varies with the use of shallower (increased distance) or more sloped (decreased distance) needle insertions. Without the need to make calculations, practitioners can directly get this distance in the long axis, offering a big picture regarding the real distance to reach the vein when using different angles of insertion (Fig. 7). Of note, knowing this distance is of paramount importance to minimize PCF, given that a large proportion of the catheter must dwell in the vein [1,6], and thus, a large distance to vein will result in a large proportion of the catheter outside the vein using standard-length catheters  (Fig. 8). This is coherent with the previous study, since 2.75 cm is closest to the 65% of an SLC. Thus, as a rule of thumb, achieving at least 2.75 cm of the catheter dwelling in the vein should be the cut-off used to mitigate catheter failure. This means, for example, that for an SLC of 4.78 cm, the "real" distance to reach the vein must be lower than 2 cm. To achieve this, several strategies can be used, for example, selecting vessels at the lowest possible depth, using sloped insertion angles, and inserting catheters which are longer than usual, such as ultra-long peripheral catheters (ULPC, 18-20G, 6.35 cm in length) and midline catheters (8-20 cm in length) [1,8,12,13]. Using catheters which are longer than the standard size aids in minimizing PFC, allowing operators to use shallower insertion angles to improve needle visualization and also to select veins which are even deeper than 16 mm. Advantages of ULPC over midline are its low costs and the fact that they do not require advanced skills such as managing the Seldinger technique, so they can be inserted by nurses or technicians.

Key concept 5: Select the cannulation technique: out-of-plane or in-plane technique
Practitioners should remember that each technique has advantages and disadvantages and thus learning and Distance from skin to vein for US-guided cannulation. In a, it is < 12 mm (ideal), while in b is > 16 mm. Since the vein in b has a diameter > 4 mm, it could still be cannulated using longer catheters than usual, for example, a midline catheter using both techniques are encouraged (Additional file 5: Video S5 and Additional file 6: Video S6), since they can select one or the other based on the situation [14,15]. In a recent systematic review and meta-analysis, greater success has been shown with the out-of-plane technique compared to the in-plane technique [16]. However, for the out-of-plane technique, the visualization of the needle tip is an important limitation, having shown a higher rate of posterior wall perforations, compared to in-plane technique, which shows a clear delineation of the needle shaft and needle tip as it is advanced from superficial tissues into the vein [17,18]. Using the "walk-down" maneuver (i.e., "follow the tip technique") improves visualization of the needle tip when using out-of-plane insertions [14] and should be considered for using in practice. Side-lobe artifact is common when performing the in-plane technique, which simulates that the needle is inserted into the vein lumen, when is in fact close to it [14]. The learning curve for the in-plane technique seems to be longer compared to the out-of-plane technique [15].

Key concept 6: Demonstrate the catheter is in the vein lumen and perform a saline flush test
After cannulating the vein, it is useful to check if the catheter is in the vein lumen, since is not infrequent that the infused solution passes easily to the subcutaneous tissue without any warning signs, thus delaying the institution of intravenous therapies. The catheter is observed as two parallel hyperechoic lines in the short, the long or both axes (Fig. 9a, b). In midline catheter insertions, the guidewire, seen as a hyperechoic linear structure, should be demonstrated before inserting the catheter (Fig. 9c). Finally, a saline flush test may be performed through the catheter, observing bubbles in the lumen vein in correctly Explaining the distance from skin to vessel in ultrasound-guided cannulation. While this distance (d) is estimated in the short axis, the real distance to reach the vein depends on the insertion angle. Assuming a 45° insertion angle, this real distance is equal to d multiplied by 1.4. Of note, the real distance decreases with sloped insertions, and increases using shallower insertions Fig. 7 Real distance from skin to vein measured directly in the long axis. As shown, shallowest insertions determine a longest pathway to reach the vein, resulting in a large proportion of the catheter dwelling outside the vein and ultimately leading to catheter failure. In contrast, sloped insertions lead to shortening the distance to reach the vein, and aid in increasing the proportion of the catheter dwelling in the vein lumen positioned catheters (Additional file 7: Video S7); color Doppler can be used for this purpose as well [19].

Conclusions
Practitioners should consider several issues when inserting intravenous peripheral catheters under ultrasound guidance, aiming to improve success rate, avoid complications and lengthen the survival of the catheter. Based on available data and everyday practice, all indicate that catheters longer than standard size are needed for USguided peripheral venous cannulation, with the purpose of minimizing premature catheter failure. This is a call for attention to catheter manufacturers, since a more affordable solution at hand is expected from them shortly.

Supplementary information
Supplementary information accompanies this paper at https ://doi. org/10.1186/s1308 9-019-0144-5.  To achieve at least 2.75 cm of the catheter dwelling in the vein, several strategies can be used, for example, selecting veins at the lowest depth possible, using sloped insertion angles, and inserting catheters which are longer than usual, such as ultra-long peripheral catheters or midline catheters Fig. 9 a, b Demonstration of the catheter (arrows) entering the vein lumen in short (a) and long axis (b); c demonstration of the guidewire entering the lumen vein (arrows) in the long axis, when using a midline catheter