Nutritional support is considered an essential part of care in critically ill patients and numerous studies have demonstrated that adequate nutrition is directly related to clinical recovery and survival.¹ Widely accepted means of nutrition are total parenteral nutrition and enteral nutrition (EN).^1–4 The latter, in the form of tube feeding, is the traditional and preferred route of nutritional support in many institutions because it is widely considered to be more physiologic and less invasive.^1,5 Despite multiple randomized trials and meta-analyses, there is ongoing controversy as to whether one is superior to the other with regard to the quality of nutrition, morbidity, and mortality.^1,2,4 Although it is still debatable in a real world setting, animal studies have suggested that with EN the mucosal integrity is preserved leading to less bacterial translocation and decreased stimulation of systemic inflammatory response.^2,3 EN is provided through a tube that can either be placed through the nostril (nasogastric or nasojejunal feeding tubes) or mouth (orogastric or orojejunal tubes) or percutaneously through the abdominal wall (percutaneous gastrostomy or jejunostomy).³ There is controversy with regard to which modality is superior.^3,6 Transnasal feeding tube is the traditional means of EN and is widely accepted due to the simplicity of its placement.^5,7 Although placement of transnasal feeding tubes is less invasive and inexpensive compared to other modalities, it has been related to poor quality-of-life measures including subjective discomfort, poor cosmesis, and influence on social activities.^6,8 Long-term feeding tube placement has also been related to the development of mucositis and paranasal sinusitis.⁹ In the clinical practice guideline published by Itkin et al,³ transnasal feeding tubes are suited for patients who require short-term enteral feeding for less than 6 weeks. Meanwhile, percutaneous gastrostomy or jejunostomy may be preferred in patients who require EN for longer periods. However, such percutaneous placement of gastrostomy or jejunostomy tubes requires more technical skill and experience and is more invasive and relative higher in cost than the insertion of transnasal tubes. Furthermore, gastrostomy or jejunostomy sites may be complicated by peristomal infection or leakage.^3,9 Transnasal feeding tubes can be placed blindly at the bedside which makes it suitable for critically ill patients who are often restrained to numerous monitoring devices or under ventilator care.^7,10,11 However, blind placement of nasogastric or nasojejunal feeding tubes is unreliable and not free from complications.^12–15 In this regard, various modalities for guidance from fluoroscopy to endoscopy have been introduced in the literature.^6,16–21 Compared to fluoroscopy, endoscopy offers additional benefits besides guidance including the ability to provide diagnostic information and freedom from risk of radiation.^18,21 On the other hand, fluoroscopy guidance is less expensive and more tolerable for the patient due to the low profile of angiographic catheters compared to endoscopic equipment. Furthermore, fluoroscopy-guided procedure does not usually require sedation which is often a necessity for endoscopic procedures.¹⁷ Equally high rates of technical success reaching over 90% have been reported for both fluoroscopy- and endoscopy-guided insertion of feeding tubes.^12,17,22 This review focuses on the technical aspects of fluoroscopy-guided placement of transnasal feeding tubes performed in the fluoroscopy or angiography unit.

Ever since the first description of enteral feeding in the sixteenth century, feeding tubes have undergone an evolutionary process with regard to the product material and design.^5,7 During this process, small bore feeding tubes made of soft synthetic material were introduced and popularized which were easier and safer to place blindly at the bedside.⁷ However, with their increasing use, a rising number of complications were reported.^12,14 The most serious complication is bronchopulmonary misplacement.^13,15 Although the incidence varies, the number of bronchopulmonary complications arising from misplacement reaches 4.0% according to the literature. Ghahremani and Gould¹⁴ have reported a misplacement rate of up to 4.4% in 340 patients who underwent blind feeding tube placement. Complications associated with misplacement include nasopharyngeal trauma, pneumothorax, hydrothorax, empyema, and pneumonia. In the study by Rassias et al,¹⁵ 0.3% of all placements directly contributed to patient death.

Various methods have been sought in order to confirm the position of the tube.^10,11 Relying on cough reflex is not recommended for patients who are either sedated or intubated for they are likely to have decreased or absent cough reflexes.¹² Visual inspection of aspirate, palpation of the epigastrium, and auscultation during air insufflation are all unreliable methods.¹¹ Meanwhile, chemical inspection (pH, enzymes, bilirubin) of aspirate, capnometry, and electrocardiogram (ECG) guidance are time consuming.¹¹ A traditional, simpler means of verification is by the use of X-ray. Fluoroscopy possesses advantages over spot radiographs taken after insertion of feeding tubes because it offers real-time monitoring during feeding tube placement.^19,20

With the increasing use of nasojejunal tubes, which have been associated with a lower risk of aspiration pneumonitis compared to nasogastric tubes,^22–24 imaging guidance is often required in order to place the tip of the feeding tubes beyond the pylorus or even into the proximal jejunum. Fluoroscopy is available in most institutions and is well-suited for such purpose because most commercially available feeding tubes appear radio-opaque on X-ray. Water-soluble contrast media helps to outline the gastrointestinal system and therefore facilitates navigation along the course of the stomach and small bowel. While blind placement of feeding tubes requires calculation of the nose-to-tip distance, with real-time fluoroscopy, the tip of the tube can be monitored while it is propagated distally and until it reaches the target location. Ott et al²² have reported a technical success rate of 90% with feeding tube during real-time fluoroscopy.

The major disadvantage of fluoroscopy is that the patients have to be transported to the fluoroscopy unit.²⁰ Critically ill patients who are under constant monitoring may be restrained to various monitoring devices and may also be under ventilator care. Complications may potentially result from moving such patients to the fluoroscopy suite. Another disadvantage is radiation exposure to the patient as well as to the medical staff.^19,20 However, fluoroscopy guidance possesses afore-mentioned advantages that outweigh its drawbacks, especially with regard to iatrogenic complications related to tube misplacement.

While conventional feeding tubes are not specifically devised for over-the-wire placement, some of the recent products (Kangaroo™; Covidien, Mansfield, MA, USA) (Fig. 1A) have been designed for this purpose and have a distal end hole through which a standard guide wire can be inserted (Fig. 1C).

Ideally, while the patients are fasted overnight before the procedure, a nasogastric tube should preferably be placed during this period in order to empty the stomach of any residual gastric contents. Not only does this prevent aspiration of gastric contents into the respiratory system during the procedure but it also helps to reduce the volume of the stomach thus preventing redundancy and unwanted looping of equipment inside the gastric lumen, particularly in the fundus. After the patient is positioned on the fluoroscopy table in supine position, the pre-positioned nasogastric tube is removed and the field around the nose is prepared in a sterile manner. Xylocaine spray is used to numb the nasal mucosa and sterile lubricant is applied inside the nostril to prevent mucosal injury that may result from friction during tube insertion. With the patient’s head slightly rotated towards the operator, the X-ray beam is rotated in the contralateral oblique direction. A combination of these two maneuvers provides a near-lateral projection of the nasopharynx which makes it easier to trace the course from the nasal cavity into the oropharynx (Fig. 2A). In this projection, discrimination of the air columns along the trachea and esophagus becomes easier.

Feeding tubes usually have a straight, blunt tip and therefore do not allow directional control during their insertion. In order to overcome this problem, a standard 0.035-inch hydrophilic guide wire with a curved tip (Fig. 1B) measuring 260 cm or longer in length is used to navigate through the tortuosity of the gastrointestinal system before the feeding tube is co-axially delivered over the guide wire. The hydrophilic guide wire often passes through tight strictures associated with benign or malignant pathologies in the oropharynx or esophagus that may be impossible to traverse with the feeding tube alone. For this reason, such over-the-wire techniques have broadened the indication of feeding tube placement for patients with proximal passage obstruction. On occasions, the curvature on the tip of the guide wire may not be sufficient to overcome severe angulations. It is therefore ideal to couple the guide wire with a 4- or 5-Fr angiographic catheter of the operator’s choice (Fig. 1B, 2B), the latter of which may be torqued to point in the desired direction. Such angled catheters are especially useful for directing guide wires along the lesser curvature of the stomach, through the pylorus, and around the duodenal bulb. Furthermore, the catheter provides stiffer support in large spaces such as the stomach. The guide wire is positioned as distally as possible into the jejunum to prevent loss of the guide wire while the feeding tube is co-axially advanced. After the angiographic catheter used for guidance is removed, the feeding tube with a distal hole is loaded onto the guide wire and advanced over the wire (Fig. 1C, 2C). Alternating “tug and release” motions on the guide wire and controlled pushing maneuvers are necessary in order to gain better support and trackability without losing the guide wire in the jejunum (Fig. 2D). Some tubes have a distal weight which facilitates distal propagation of the tube (Fig. 1C, 2C). Using the over-the-wire technique under fluoroscopy guidance, Hwang et al²⁰ reported a technical success rate of 100% in their group comprising infants and children without occurrence of major complication. The mean procedure time was 10.6 ± 6.8 minutes.

Conventional nasogastric or nasojejunal tubes, such as the Levine tube, are intended for bedside placement and therefore lack an end hole in the distal tip. Instead, these tubes have large side holes that allow drainage and administration of nutritional formula, medication, and fluid. Without modification, the guide wire can only be passed through the distal side hole which results in diversion of the distal tip of the tube from the course of the guide wire. Such configuration occasionally becomes problematic when the tube tip is caught in angulated anatomy or mucosal folds in the duodenum. To overcome this, a hole can be cut at the tip using scissors or drilled into the distal tip using a 16-gauge needle (Fig. 3A). Distal holes that have been created in this fashion suffice to allow passage of a regular 0.035-inch guide wire (Fig. 3B). When the nasojejunal tube is co-axially advanced over the guide wire, friction may occur between the guide wire and inner wall of the tube lumen due to the long length of the tube and due to the lack of hydrophilic coating along the inner surface of the tube lumen. Park et al²⁵ devised a technique to overcome this limitation during the placement of Miller-Abbott tubes which is used for bowel decompression in patients with small bowel obstruction. With their technique, an angiocatheter is used to puncture the tube 50–60 cm proximally from its tip. After removing the metallic stylet from the angiocatheter, the guide wire is passed through the end hole of the Miller-Abbott tube and externalized through the angiocatheter resulting in a system that morphologically resembles a monorail system. Levin tubes can be modified in the same manner in order to reduce friction between the guide wire and inner surface of the tube (Fig. 3C). Furthermore, the operator is not restricted by the limited length of the guide wire should there be need to exchange catheters during the procedure. In their study, Park et al²⁵ compared their new technique with the conventional technique where the Miller-Abbott tubes are placed without the use of a guide wire. There were no technical failures with their new method, while a technical success rate of only 40% was achieved without a guide wire. Technical failures owed from failure to navigate through the pylorus or the transitional segment between the duodenum and jejunum. In addition, the procedure time was significantly shorter with the use of a guide wire. Their technique can be adopted for the placement of Levine tubes (Fig. 4) to decrease failure rate and procedure time.

Tube feeding remains an essential means of providing nutrition support to critically ill patients. Compared with traditional bedside techniques, the combination of fluoroscopic guidance and over-the-wire technique decreases the rates of technical failure and complications associated with misplacement.