According to the working principle of shunt tubes [ 12 ], intracranial pressure (ICP) = ventricle + peritoneal hydrostatic difference (HPD) + valve opening pressure (VOP) + peritoneal pressure (IAP) when using adjust-able pressure tube shunt; and valve opening pressure (VOP) = intracranial pressure (ICP) difference between ventriculoperitoneal hydrostatic pressure (HPD) - anti-pressure device setting pressure (GD) - peritoneal pressure (IAP). The fluid that is normally shunted into the peritoneal cavity is easily absorbed. Therefore, the ventriculoperitoneal shunt model is considered to be an open system, and the cerebrospinal fluid entering the abdomen does not affect the intraperitoneal pres-sure. Consequently, the influence of hydrostatic pres-sure at the end is not considered in pressure regulation. However, when the end of the shunt duct in the abdom-inal segment is locally wrapped to form a local sac, the original open system becomes a limited sac system, and the original abdominal pressure is replaced by the pressure of the limited sac system, resulting in improving the hydrostatic pressure at the end. Therefore, the actual intracranial pressure is greater than the open cranial pressure, resulting in shunt insufficiency. In the early stage, the symptoms may be temporarily relieved by lowering the shunt valve pressure, but when the local cystic pressure at the end of the shunt peritoneal cavity is further increased, exceeding the minimum shunt pressure difference, the shunt no longer works properly. The local adhesion also makes the outlet of the shunt tube unable to change position. In general, the position of the end of the shunt tube in the abdominal segment changes with intestinal peristalsis ( Fig. 1A, C, D ), and motility at the end of the shunt tube is impaired due to local adhesion or wrapping ( Fig. 1B ). At the same time, most shunt tubes can be developed on X-ray. Based on this, we hypothesized that in patients with high suspicion of shunt insufficiency due to wrapping or abdominal adhesion, the diagnosis can be made by the DPAR showing a relatively constant local position of the shunt tube end in the abdominal segment.

A total of 82 patients diagnosed with postoperative complications of VP performed for hydrocephalus who were admitted to our unit for surgical intervention from April 2017 to December 2023 were enrolled. All included patients provided signed informed consent to participate in the study.

The patient sample selection procedure is shown in Fig. 2 . After screening according to inclusion and exclusion criteria, 30 cases with DSI were included, as shown in Fig. 2 . Inclusion criteria were as follows: (1) patients with hydrocephalus who received VP shunt more than 1 week prior, with suspected shunt disorder; (2) clinical symptoms were obviously improved initially, and then became worse, such as headache or consciousness disorders and other neurological dysfunction; (3) CT or MRI showed ventricle enlargement or cerebral edema; (4) adjustable pressure shunt tube was adjusted to the lowest level, but without improvement; (5) the physical examination revealed that the shunt tube fluid storage bag exhibited good elasticity, the continuity of the shunt was not significantly compromised, and there was no evidence of local effusion in the subcutaneous tract of the shunt; (6) all men and non-pregnant women.

Exclusion criteria were as follows: (1) complicated with intracranial infection; (2) complicated with intracranial hemorrhage; (3) obstruction at the end of the ventricle; (4) shunt infection; (5) shunt valve failure; (6) shunt tube continuity disorder, such as rupture or tie or separation at the junction; (7) peritoneal cyst at the end of the shunt tube; (8) peritoneal infection; (9) shunt tube dependence syndrome; (10) excessive shunt; (11) shunt tube exposure; (12) shunt tube failed to develop on X-ray.

All patients were examined by DPAR with X-ray equipment (SHIMADZU Corp. Kyoto, Japan). The X-ray tube operated at 65 kV (peak) voltage, the tube current was 320 mA, and the X-ray volume was 51.2 mAs.

Imaging assessments were conducted independently by two radiologists each with 8 years of experience on standard PACS workstations. In order to facilitate proper abdominal peristalsis for the propulsion of potential shunt ends, DPAR was performed again 4-6 h (A working time zone) after the first radiography.

Changes in position at the end of the shunt tube in the abdominal segment were compared, and the distance values of the position changes were measured, as shown in Fig. 3 . The initial step in the measurement process involved identifying and marking the midpoint (point A) of the inferior border of the sacroiliac joint junction line (line A) on the first plain abdominal radiograph. Subsequently, a vertical line (line B) was extended from point A, followed by the drawing of another vertical line (line C) connecting the end of the abdominal segment of the shunt (point C) to line B. The point of intersection between line B and line C was subsequently identified and marked as point B. The distance between point B and point C was then measured, as illustrated in Fig. 3A . Based on the parameters and methods assessed in the initial plain abdominal radiograph, point C was identified on the subsequent plain abdominal radiograph 6 h later. Subsequently, the location of the end of the abdominal segment shunt on the subsequent plain abdominal radiograph was designated as point C’, and the linear distance between points C and C’—referred to as line D—was determined and recorded (see Fig. 3B ). The PACS workstation was utilized for objective measurement utilizing its proprietary software. To mitigate the impact of varying angles in the visual angle collection on judgment results, three fixed reference points and three standard lines were established during the measurement process. This allowed for the evaluation of the distance between the end of the shunt tube and the three lines. The resulting values were averaged based on independent measurements conducted by two assessors.

Abdominal exploration was performed, and the shunt distal tubing was adjusted. After dredging the abdominal end of the shunt catheter, the cerebrospinal fluid drainage was examined to ensure it was unobstructed; another small incision was made in the abdomen to re-insert the abdominal end of the shunt catheter into the abdominal cavity. Since the adhesion area at the end of the shunt is a limited space formed by the greater omentum wrapping, and this wrapping can be around the channel formed by the shunt to the abdominal wall incision, a new incision must be made 3CM away from the original incision when adjusting the shunt to avoid entering the limited space along this channel when re-catheterizing. Abdominal shunt adjustments performed in accordance with standard postoperative care do not require specialized care.

A neurosurgeon with 10 years of experience evaluated each patient’s clinical data, including hydrocephalus etiology, symptoms, clinical examination, intraoperative findings, CT, and X-ray images.

Verification criteria were as follows: an exploratory operation was performed at the peritoneal end of the shunt tube to determine whether the exit of the peritoneal end was obstructed, resulting in poor cerebrospinal fluid drainage of the shunt tube.

Efficacy judgments were as follows: (1) effective, improvement of clinical symptoms—headache, vomiting, disturbance of consciousness, and other clinical symptoms improved or Glasgow Coma Scale (GCS) score increased. Imaging improvement, cranial CT showed relief of ventricular pathologic dilatation or cerebral edema. If the first two items are not satisfied, changes in the position of the shunt outlet in the abdominal section are shown by the dynamic abdominal plain film. (2) Ineffective, no improvement in clinical symptoms or CT imaging.

Categorical variables, including gender, clinical symptoms, etiological frequency, and percentage, are presented as numbers and percentages. Quantitative data are expressed as the mean plus standard error (SEM) of the mean. Graphic images were generated on a bioinformatics platform. Statistical analysis for multiple group data was performed using the Analysis of Variance (ANOVA) method followed by the Students t-test or Wilcoxon rank sum test to compare data between the two groups. Differences in EVAN scores and GCS scores before and after surgery, as well as differences in the position of the end of the shunt tube in the abdominal cavity between the under-shunt group before surgery and the negative patency group after catheterization, were analyzed. Statistical analyses were performed using the GraphPad Prism (version 9.1.0, San Diego, CA), and P < 0.05 was considered statistical significance.

The study included 30 patients, including 16 males (53.3%) with a mean age of 42.1 ± 24.7 years, a range of 9-82 years; and 14 females (46.7%) with a mean age of 56.9 ± 16.3 years, a range of 18-79 years. Shunt insufficiency varied between patients, occurring from 2 weeks to 20 years after VP shunt surgery (mean 2.95 ± 5.09 years). In 20 cases, shunt inefficiency occurred within 2 years after the VP shunt. Based on the classification of hydrocephalus etiology, the times of shunt insufficiency after the first VP shunt are shown in Fig. 4 . DSI occurrence times of unacquired hydrocephalus after the first VP shunt were longer than those of acquired hydrocephalus ( P < 0.01), suggesting that the problem of under-shunt of secondary hydrocephalus in the short term requires extra attention. Patients’ clinical characteristics are shown in Table 1 .

DPAR of 30 patients showed that the abdominal end of the shunt catheter was fixed ( Fig. 5B, E ), and the shunt end adhesions caused by the shunt catheter were diagnosed as shunt insufficiency.

All patients underwent a second VP shunt operation. In 28 cases, the abdominal end catheters were pulled out and a small amount of retinal fats or cellulose membrane-like tissues were wrapped at the end. Fibrin-like tissue blocked the end of the shunt catheter in one case. During the operation, local adhesions or blocked tissues were removed, and cerebrospinal fluid (CSF) was automatically drained out, or CSF was automatically drained out after the fluid storage sac was pressed again. The automatic drainage of CSF was observed to be smooth, and the shunt catheter was re-inserted. In the other one case, no tissue blockage and no improvement in postoperative symptoms and imaging were observed, for which the cause was finally confirmed to be poor ventricular drainage, and which was cured after VP with a shunt catheter inserted in the other ventricle. After the second VP shunt operations, CT imaging showed the patients’ ventricles to have become smaller, indicating adequate drainage ( Fig. 5D ). The DPAR showed changes in the position of the shunt catheter at the abdominal end ( Figs. 5C, F ). The diagnostic accuracy of the DPAR was 96.67% (29/30). Comparisons were made between the positive group of preoperative shunt tube terminal position changes and the negative group of postoperative shunt tube patency. The end of the shunt tube in the abdominal segment of the preoperative group was fixed with DPAR at two different times, and the change was 1.74 ± 1.18 cm. Compared with the postoperative and negative groups, the change in the position of the end of the shunt tube in the abdominal section at different times was 9.36 ± 2.64 cm, which showed a significant difference compared with changes in the preoperative group ( P < 0.001) ( Fig. 6 ).

Patients’ follow-ups ranged from 3 months to 2 years. The clinical symptoms of 20 patients were improved, while the clinical symptoms of 3 patients were not improved, but the imaging was improved. The postoperative Evans index (0.37 ± 0.08) was significantly lower than the preoperative Evans index (0.42 ± 0.08) ( P = 0.007), and the GCS score (12.8 ± 2.69) was significantly higher than preoperative scores (11.36 ± 2.43) ( P = 0.013) ( Fig. 7 ).

DPAR is a simple and effective method to identify shunt insufficiency caused by delayed abdominal adhesion in patients with hydrocephalus after VP surgery.