Feasibility and safety of a novel electromagnetic device for small-bore feeding tube placement

Study design

This prospective, single-arm study was conducted in the intensive care units of two institutions in the USA: the Cleveland Clinic Foundation (CCF), Cleveland, Ohio and St. Vincent Indianapolis Hospital and Health Care Center (St. Vincent), Indianapolis, Indiana. Patients who met study eligibility criteria were enrolled during December 2017–March 2018 at CCF, and during April 2018–August 2018 at St. Vincent. Written informed consent was obtained from the patients or their legal authorized representative prior to study participation.

The placement team at CCF consisted of 3 nurses and 2 physicians; and at St. Vincent, 10 nurses. All clinicians involved in this study were previously trained in the placement of small-bore feeding tubes as per institution requirements.

All clinicians received formal training on the EMTS by a company representative. A dedicated session consisted of an overview of the system components and EFT, system set-up and EMTS image interpretation. Training also included a return demonstration of proper procedure on a mannequin with airway and GI structures that enabled visualization of airway misplacement and correct gastric and postpyloric placement. Each nurse completed one to two demonstrations to ensure competency operating the system. Moreover, a representative of the company was present to observe each placement and to ensure safety and proper use.

Patient selection

Study inclusion criteria were age 21 years or above, requiring placement of an EFT; who either had an endotracheal tube (ETT), or did not have an ETT but were sedated (Richmond Agitation-Sedation Scale of −2 or less) or obtunded (Glasgow Coma Scale (GCS) of 9–12). Eligibility for participation in the study also included the ability of the patient or legally authorized representative to understand and adhere to all protocol procedures and to provide written informed consent.

Exclusion criteria included a history of esophageal varices or ulcers, upper GI stenosis or obstruction, upper airway obstruction, trauma involving the sinuses, the nares, the face or the neck that would preclude EFT insertion in the oral or nasal route, deformities of the sinus cavities or skull base, esophageal cancer or neoplasm, a significant concomitant illness that would adversely affect participation and pregnancy or lactation.

Electromagnetic tracking system

All EFTs were placed with the aid of the EMTS. The system is composed of a dedicated mobile cart and is equipped with a battery-operated all-in-one computer. The computer runs the software, which controls the EMTS. The system uses a field generator and several EM sensors, which enable proper scaling of the display to the patient’s body contour. The EFT was designed to operate specifically with the EMTS. The EFT contains an EM sensor incorporated into its distal tip, enabling continuous position confirmation, and creating a path on the monitor. The EMTS field generator produces a series of varying magnetic fields that create a known sensing volume of varying magnetic flux. The position and orientation of the distal tip of the EFT is tracked by the EMTS and the system displays the placement pathway. This pathway is shown on the monitor in three views: frontal, lateral and axial, to allow for a more accurate three-dimensional interpretation of placement. The frontal view represents the tube tracing looking towards the patient. The lateral view represents a side view of the patient and depth of the tube relative to the patient’s chest. The axial view is the transverse view of the patient looking from foot to head. The system does not require placement of a device on the patient’s body as the EM field generator is located on the system cart, which is placed at bedside. Additionally, as the EM sensor is embedded within the tube tip, the EMTS enables re-confirmation of the EFT tip location by simply reconnecting the EFT to its cable.

The marking of the two anatomical landmarks: (1) suprasternal notch and (2) xiphoid process using a registration pen, together with the positioning of the reference sensor on the midaxillary line enables scaling of the display to the patient’s upper body outline. An ideal division of the frontal screen on the EMTS into four quadrants is created by means of a crosshair. In contrast to the Cortrak receiver, which must be placed on the patient’s abdomen during the entire insertion procedure, the origin of the crosshair on the ENvue EMTS screen is determined only once during the initial set-up by marking the xyphoid process and remains stable and accurate throughout the subsequent procedure even with patient movement. With the Cortrak system, any movement of the receiver may impact the display grid and result in an inaccurate tracing.

With the ENvue EMTS, the two upper quadrants indicate the area of the lungs. The xiphoid process (lower anatomical landmark) represented by the central crosshair of the grid, specifies the transition from the esophagus to the stomach, with the left lower quadrant containing the stomach area and the right lower quadrant the area of the duodenum. The vertical line of the crosshair below the horizontal axis marks the area from which the stomach transitions to the pylorus and beyond to the duodenum and small intestine. The EFT is deemed MRI conditional per FDA.

Study protocol

Patients were screened to ensure compliance with the study inclusion and exclusion criteria. Following enrollment, baseline data were collected, and a physical examination was performed. All patients were placed in supine or semi-Fowler’s position for the start of the placement. All EFTs were placed using the EMTS at bedside according to the manufacturer’s instructions. The clinician performing the placement selected one of two sizes of EFTs: 10 Fr, 140 cm or 12 Fr, 140 cm, according to the patient’s clinical needs; the protocol for placement was the same for both size options. The EMTS monitor with visualization of the insertion tracing was used for guidance throughout placement. Following tube placement, an abdominal radiograph was obtained to verify tube location. Tube tip location was then verified by the EMTS and by radiography at 12–49 hours following placement (at a follow-up visit). EFT tip locations per EMTS and radiograph were adjudicated and compared by a specialist in GI medicine. The GI specialist, who was blinded to the procedure, interpreted tip location according to EMTS and radiograph. The EMTS insertion tracing and radiograph of the same procedure were not viewed at the same time. The study coordinator then assessed agreement between the EMTS and radiographs regarding placement within the GI tract. For the EMTS and radiograph results, the tip location was designated as: small intestine (duodenum or jejunum), stomach, pylorus region, lung or ‘other’. If ‘other’, a specific location was recorded. Vital signs were collected prior to tube placement and during the follow-up visit. Occurrences of any adverse events (AEs) or device-related side effects were also recorded.

Outcomes

The primary outcome of the study was the placement of the EFT in an appropriate anatomical position, that is, within the GI system, without the occurrence of guidance-related AEs, such as pneumothoraces. Correct placement was determined by comparing tube tip location observed by the EMTS with an abdominal radiograph. This comparison verified that there was no mismatch between the EFT location as observed on the screen of the EMTS and the radiograph. A critical mismatch was defined as a scenario in which the EMTS would show placement in the GI tract, while the EFT was observed in the airways according to the radiograph. Any guidance-related serious AEs (SAEs) or AEs occurring during patients’ participation in the study were evaluated and documented.

The secondary outcome was the accuracy of re-confirmation of tube tip location using the EMTS. This was evaluated 12–49 hours after the procedure (at a follow-up visit), by re-connecting the EFT to the EMTS and comparing the EFT tip location on the screen with that observed on radiography. Timing varied based on patient condition and/or scheduling needs. In addition, the possibility of retrograde tube migration from the initial location was examined, and safety assessments that were not guidance-related were performed. Safety of the EMTS was evaluated by assessment of vital signs (pulse, respiration, blood pressure and oxygen saturation) during the procedure visit and during the follow-up visit. Non-guidance-related SAEs and AEs occurring during study were recorded.

Data collection

The study database was designed and managed using Medrio Electronic Data Base (Medrio, San Francisco, California, USA). Following patient enrollment, baseline data were collected; these data included patients' demographics, weight, height, medical history, concomitant medications and the presence of an ETT. If a patient did not have an ETT, the level of sedation or obtundation was evaluated. Prior to tube placement, a physical examination was performed, and vital signs were recorded. These data were also recorded prior to the follow-up re-confirmation of tube tip location. Additional variables related to the EFT placement procedure were documented, such as tube size and insertion route (oral or nasal). The occurrence of AEs and device-related side effects during the study were also recorded.

Study population

Seventy consecutive patients in total were enrolled (22 at CCF and 48 at St. Vincent). Five patients were excluded from the study analysis. Reasons for exclusion are shown in figure 1. The remaining 65 patients were included in the intent-to-treat analysis and underwent a tube placement attempt. Of these, 57 (88%) completed a successful placement procedure and 8 did not (see reasons in figure 1). Of the 57 patients who completed a successful placement procedure, 48 (84%) underwent a successful tube tip position re-confirmation at the follow-up visit. Two patients pulled the EFT out and a new tube was placed several hours later in one of them. In total, 58 tubes were placed in 57 patients. Twenty-five procedures were performed using a 12 Fr tube and 33 with a 10 Fr tube. Of the successful tube placements, 31 (53.4%) were oral and 27 (46.6%) nasal.

Demographic characteristics and baseline data of the patients included in the analysis, according to successful placement are presented in table 1. Patients’ age ranged from 23 to 88 years (median age 64 years). Twenty-four (37%) were females and 41 (63%) males. All patients had an ETT except for six patients who were obtunded (two patients with GCS of 9, two with GCS of 10 and two with GCS of 11). Placement was successful for all the obtunded patients.

Primary outcome

For all 58 successful placements, an EFT was appropriately placed in the GI tract, according to radiography. According to both the EMTS and the radiographic confirmation (figure 2), there were no lung placements and no pneumothoraces or any other guidance-related AEs. In two patients, feeding tube deviation from midline (vertical line of crosshair) above the horizontal axis was noticed on the EMTS screen suggesting EFT passage towards pulmonary structures. This was recognized by the clinician and EFT was removed promptly. Radiography showed 27 tubes placed in the small intestine, 22 tubes in the stomach and 9 tubes in the pylorus region. Complete agreement of tube tip location between the EMTS evaluations and radiographs was achieved for 56 of the 58 placements: 22/22 in the stomach, 9/9 in the pylorus region and 25/27 in the small intestine (shaded values in table 2A). One patient was described by the EMTS as having the distal feeding tube tip in the pylorus region and by the radiograph as being positioned in the gastric antrum, or first portion of the small intestine. The location reported by the EMTS is in between the two locations described by the radiograph; therefore, the placement was categorized in the pylorus region according to both the EMTS and the radiograph. Two patients were described by the EMTS as having the distal feeding tube tip in the stomach region and by the radiograph as being in the small intestine.

Secondary outcomes