Direct, real-time visualization-assisted nasoenteral tube placement in critically ill patients: a multicenter experience

To the Editor:

Postpyloric feeding is recommended for ICU patients with gastric feeding intolerance or a high risk of aspiration [1]. Timely enteral nutrition depends on rapid and accurate tube placement. Although endoscopic placement is one of the most reliable methods [2], its dependence on endoscopist availability and specialized equipment may delay feeding. Real-time visualization-assisted tube placement (VATP) uses a micro-camera integrated into the tube or guidewire tip to provide intraluminal visualization, enabling ICU caregivers to place the tube under direct view. Given the potential influence of operator experience on procedural outcomes [3], we report a prospective multicenter experience from five centers in China, describing the feasibility and learning characteristics of VATP in ICU practice.

The VATP system we used has been described previously [4]. It consists of a feeding tube, a reusable guidewire with a micro-camera at its tip, and an image processor (Jiangsu Jianzhiyuan Medical Device Technology, China). The guidewire was inserted into the feeding tube and connected to the processor for real-time visualization. ICU nurses and clinicians without prior VATP experience completed a training program including anatomical landmark recognition, expert demonstrations, and supervised hands-on practice, followed by prospective consecutive first-attempt placements. A standardized placement protocol was used across centers. The tube was advanced until intestinal villi were seen, indicating intestinal entry (endoscopy-defined success, EDS) (Fig. 1). Attempts could be stopped by the operator or supervisor before the maximum permitted duration of 2 hours. Success was defined as EDS followed by X-ray confirmation that the tube tip was positioned in the first part of the duodenum (D1) or beyond.

Fig. 1.

Visualized tube placement pathway and corresponding anatomical landmarks

Twelve operators placed 316 procedures. We summarized 308 placements performed by the 10 operators who completed at least 30 procedures for feasibility and learning curve assessment. Villi were visualized in 294 placements, and radiographs confirmed postpyloric positioning in 292 (94.8%). In two EDS cases, the tube had slipped back before radiography, both after only a short intestinal advance. The tip reached fourth part of the duodenum (D4) or beyond in 199 (64.6%). Individual success rates ranged from 82.4% to 100%, with 8 of 10 operators exceeding 90% (Table S3). The operator with the lowest success rate had six failures, all within the first 20 cases. Median total procedural time was 25 min (IQR, 15–37.5), including 15 min to villi (IQR, 8–25) and 7 min thereafter (IQR, 4–12). Further details are provided in the Additional file.

VATP provides a clear, real-time visual endpoint; however, the thin, flexible tube offers limited active steering, and initial unfamiliarity may prolong earlier procedures. We therefore assessed the learning curve through procedural efficiency. Procedural times were ordered chronologically, and each procedure’s deviation from the operator’s mean time was cumulatively summed to generate the CUSUM curve [5]. A cubic polynomial was fitted, and the point where the slope changed from positive to negative defined the breakpoint. A breakpoint was identified in seven operators at a median of 20 procedures (range 13–26) (Fig. S1). For descriptive comparison, procedures were grouped as an early phase (cases 1–20) and a late phase (cases ≥ 21).

Median total procedural time decreased from 27 min (IQR, 17–42) in the early phase to 20 min (IQR, 13–30) in the late phase (Table 1). After accounting for operator-level clustering, the adjusted GMR was 0.77 (95% CI, 0.66–0.89). Radiographically confirmed success was similar between phases (95% vs. 94.4%). By contrast, distal placement at or beyond D4 increased from 60.5% to 72.2% (adjusted OR, 1.73; 95% CI, 1.03–2.89).

Table 1.

Comparison of procedural performance between early (cases 1– 20) and late (cases ≥ 21) phases

Performance	Cases 1–20 (n = 200)	Cases ≥ 21 (n = 108)	Unadjusted P value	Adjusted estimate (95% CI)	Adjusted P value
Radiographically confirmed success	190 (95)	102 (94.4)	0.83	OR 0.96 (0.33– 2.79)	0.94
Placed at D4 or beyond	121 (60.5)	78 (72.2)	0.04	OR 1.73 (1.03– 2.89)	0.037
Procedural time, min	27 [17– 42]	20 [13– 30]	< 0.001	GMR 0.77 (0.66– 0.89)	< 0.001
Airway entries	4 (2)	2 (1.9)	1	–	–

Data are presented as number (percentage) for categorical variables and median [IQR] for continuous variables. Unadjusted P values were obtained from between-group comparisons using the Mann-Whitney U test, chi-square test, or Fisher exact test, as appropriate. Adjusted OR for binary data and adjusted GMR for time data were estimated using mixed-effects models with operator included as a random intercept; time outcomes were log-transformed. OR, odds ratio; GMR, geometric mean ratio; IQR, interquartile range. No adjustment for multiple comparisons was applied; P values should be interpreted as exploratory

Safety was assessed in all 316 procedures. Overall, the procedures were well tolerated, with no severe adverse events. Minor events were transient, including vital sign alerts (4.4%), epistaxis (4.4%), mucosal oozing (3.8%), and agitation (3.2%). Airway misplacement occurred in 6 cases (1.9%), four within the first 10 cases and two after case 25. Although VATP cannot completely prevent airway entry, all cases were detected in real time and corrected.

In this multicenter experience, VATP-assisted postpyloric tube placement by ICU caregivers was feasible, with high placement success rates. Procedure time decreased with increasing experience, suggesting improved procedural efficiency. The anatomical landmarks were readily identifiable, supporting the potential incorporation of video-based materials and hands-on practice into future training. In addition, VATP can provide a practical bedside option when endoscopic resources are limited, while endoscopy remains an important alternative or rescue option when bedside placement is unsuccessful or more definitive control is required.

Consecutive cases were prospectively recorded, reducing the likelihood of selective case inclusion. However, variation in case mix across procedures should be considered when interpreting these findings. The use of a single guidewire-based system and the limited number of procedures per operator also restrict the assessment of the learning curve. The median procedure time was 25 min, and EDS was achieved within 50 min in 286 procedures (92.9%) in our cohort, suggesting that a shorter practical time window may be appropriate.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

VATP

Visualization-assisted nasoenteral tube placement

EDS

Endoscopy‑defined success

CUSUM

Cumulative sum

D1

The first part of the duodenum

D4

The fourth part of the duodenum

Author contributions

Jun Jin, Xin Tian and Yifeng Luo conceived this experience and critically revised the manuscript for important intellectual content. Yifeng Luo had full access to all of the data. Guoli Yang, Jinhan Huang, Huaqin Wang, and Shaozhen Chen contributed to the training and placement protocol. Guoli Yang performed the final statistical analyses and validated the results. Hui Yi and Guoli Yang contributed to data interpretation and wrote the original manuscript. Yihao Liu and Qian Zhou provided statistical expertise and quality control assistance. All authors collected data at their centers and accepted responsibility for the integrity and accuracy of the work. Kar Keung Cheng critically revised the manuscript for important intellectual content. All authors have reviewed and approved the final version of the manuscript.

Funding

This work was supported by Jiangsu Jianzhiyuan Medical Device Technology Co., Ltd., which paid a contract research organisation for the Electronic Data Capture (EDC) system, preliminary data processing and statistical analyses. No payments were made to the authors or their institutions. The funder had no access to the data and no role in drafting this report. The authors were responsible for data collection, statistical analysis, interpretation, and manuscript writing.

Data availability

Data available upon reasonable request. Proposals should be addressed to Yifeng Luo, lyif@mail.sysu.edu.cn.

Declarations

Ethics approval and consent to participate

The work was approved by the Institutional Review Board of The First Affiliated Hospital of Sun Yat-sen University (Approval No. [2023]846-1) and the ethics committees of all other participating centers.

Consent for publication

Written informed consent was obtained from the patients or their legally authorized representatives.

Competing interests

The authors declare no competing interests.