This study is a retrospective analysis of patients who underwent surgery for oesophageal cancer at our hospital between 2021 and 2024. All included patients subsequently developed a cervical anastomotic leak (CAL) and were treated with our modified NPWTi protocol. Patient data, including healing days, treatment frequency, comfort level, wound odour, and patient satisfaction, were collected.

A diagnosis of CAL was based on one of the following:

The primary and secondary outcomes in this study were as follows:

Data were statistically analysed using IBM SPSS Statistics 26.0. Normally distributed continuous variables are presented as mean ± standard deviation; non-normally distributed variables as median and interquartile range; and categorical data as number (percentage). Wound odour scores, being ordinal data, were compared using the Wilcoxon signed-rank test. A P value of <0.05 was considered statistically significant.

Tension, local blood supply, and surgical duration are recognised risk factors for CAL after oesophageal cancer surgery. Additional perioperative factors include neoadjuvant therapy (1, 6), diabetes, body weight (7), and preoperative hypoproteinaemia. Conventional treatment for CAL involves complete opening of the cervical incision, adequate drainage, and daily wound cleaning until healing. Patients typically require fasting, enteral feeding tube placement, nutritional support, and antibiotics. This treatment process not only has a long duration but also increases hospital stay and costs, while frequent dressing changes add to the workload of healthcare staff and create psychological stress for patients.

NPWT is a recent therapeutic approach that promotes wound healing. Continuous negative pressure removes wound exudates and sources of infection, prevents the spread of infection, and protects the wound. NPWT promotes neovascularization and the expression of related growth factors (8–10). It is widely used to treat pressure ulcers, leg ulcers, poor postoperative healing, diabetic foot ulcers, and various wounds.

Compared to NPWT, NPWTi enhances wound hydration through a stable irrigation fluid, accelerates wound cleaning, and aids in the dissolution and removal of deep necrotic tissue, thereby reducing the absolute bioburden (11) and improving bacterial clearance. NPWTi significantly reduces healing time for infected wounds compared with that of traditional moist wound healing (29.6 ± 6.5 vs. 13.2 ± 6.8 d, P < 0.001) (12).

However, healing in CAL cases is fundamentally characterised by the growth of granulation tissue, forming a sinus with prolonged exposure to digestive fluids and saliva, leading to slow healing (13). A meta-analysis of 11 studies indicated that the average duration of CAL treatment was 34 days (14). In this context, the primary outcome of our study demonstrates that the use of our modified NPWTi was associated with a substantially reduced time to healing of 15.00 ± 7.26 days. This finding is particularly notable when considering the expected healing timeline with standard care. While previous studies have confirmed that NPWT is effective and well-tolerated in CAL treatment (15), the risk of infection dissemination with standard NPWTi remains a concern, which our modifications aimed to mitigate.

During surgery, the anatomical positioning of the oesophagus allows the cervical incision to connect with the mediastinum and thoracic cavity. Continuous infusion and irrigation are required to maintain an effective negative pressure and avoid dispersion of the irrigation fluid. However, excessive suction may hinder the effectiveness of NPWTi, necessitating a balance between the infusion and suction. We propose the following modifications:

As secondary outcomes, wound odour management and patient comfort were areas of significant focus. We utilised Grocott’s wound odour assessment criteria to evaluate patients before and after receiving the modified NPWTi, revealing notable improvements. NPWTi effectively seals the cervical incision, isolating it from the odours of digestive fluids and saliva, a result that was well received by most patients and their families. Furthermore, patients reported comfort during the treatment, with a maximum pain VAS score of 3, and expressed overall satisfaction with the procedure.

In our study of 17 patients, 14 required only one session of modified NPWTi, whereas three required two sessions, primarily because of incomplete removal of necrotic oesophageal and gastric tissue post-CAL diagnosis. After initiating NPWTi, the nursing staff regularly cleaned the drainage bottles and monitored changes in the exudate without requiring additional treatment for the cervical incision, significantly reducing the workload.

This study has several limitations that should be considered. Firstly, it is a single-centre exploratory investigation with a small sample size and the absence of a concurrent control group, which limits the generalisability and statistical power of its findings. Future validation through prospective, large-scale, multicentre studies is therefore necessary. Secondly, our method’s reliance on the hospital’s central negative pressure system restricts patient mobility. There is a practical trade-off between the wide availability and lower cost of central systems and the mobility and potentially more precise control offered by portable negative pressure devices, which come at a higher cost. This consideration directly impacts resource use and implementation strategy in different clinical settings (15). Furthermore, while the assessment of secondary outcomes like odour improvement was performed using established criteria, the perception of odour remains somewhat subjective. Future research should aim to develop and apply more objective, quantitative assessment tools to minimise potential bias.

Despite these limitations, the modified NPWTi demonstrates promise as a feasible and safe intervention for managing CAL. It is associated with a potentially shorter healing time compared to literature-derived benchmarks for standard care, alongside observed benefits in symptom control and nursing efficiency. These encouraging findings warrant further investigation through robust comparative studies, which are needed to validate the efficacy and cost-effectiveness of this modified approach and to guide its broader clinical implementation.