Spontaneous Umbilical CSF Fistula Due to Migration of the Peritoneal End of VP Shunt: A Case Report and Review of Pathogenesis.

Abdominal complications after ventriculoperitoneal (VP) shunt surgery for hydrocephalus have been known to occur. The more common complications include peritoneal pseudocyst, obstruction of the lower end, and shunt infection. Perforations of the intra-abdominal hollow viscera leading to spontaneous extrusions of the peritoneal catheter via the natural orifices have also been reported. A rarer phenomenon still is the migration of the lower end of the VP shunt through the anterior abdominal wall, leading to the formation of a spontaneous umbilical fistula at a site unrelated to the surgical site. Eight cases have been described in the literature so far with various causes elucidated. We report this condition in a child 4.5 years after his shunt surgery and postulate different mechanisms for both early and late presentations of this condition.

INTRODUCTION

Ventriculoperitoneal (VP) shunting can present with abdominal complications such as peritoneal pseudocyst, obstruction of the lower end, and shunt infection. Uncommonly, hollow viscus microperforations can occur due to constant pressure of the shunt tip. This may result in shunt extrusions through the natural orifices. Rarer still is the migration of the lower end of the shunt through the anterior abdominal wall to the umbilicus, resulting in a spontaneous umbilical cerebrospinal fluid (CSF) fistula. We describe a 5-year-old male who underwent VP shunt at 6 months of age presenting with an umbilical fistula that was managed conservatively with good results.

CASE REPORT

A 5-year-old male child who underwent right VP shunt surgery for tuberculous meningitis with hydrocephalus at the age of 6 months presented to us with a watery discharge from the umbilical region for 15 days. He had completed his course of antitubercular therapy and was otherwise asymptomatic. He was neurologically intact and had no signs of meningitis or systemic infection. Local examination of the umbilicus revealed a sinus opening with pouting granulation tissue discharging CSF [Figures 1 and 2]. The surrounding skin was inflamed and indurated and the abdominal end of the shunt was indurated and tender. There was no other scar present on the anterior abdominal wall and no evidence of an umbilical hernia. A noncontrast computerized tomography scan (NCCT) of the brain revealed the ventricular end of the shunt in situ with arrested hydrocephalus [Figure 3]. The ventricular end of the shunt was removed by making a small incision in the scalp and was disconnected from the lower end. A horizontal incision was made over the lower end of the abdomen and the shunt was removed. CSF was seen leaking from the sinus opening. In the postoperative period, the patient was observed for any signs of raised intracranial pressure, and a repeat NCCT scan showed no increase in the ventriculomegaly. The child was discharged after 7 days, and on follow-up at 3 months, the child is doing well.

Figure 1

Clinical photograph showing umbilical adenoma with purulent discharge

Figure 2

Clinical photograph showing cerebrospinal fluid leak

Figure 3

Noncontrast computerized tomography head showing arrested hydrocephalus with ventriculoperitoneal shunt in situ

DISCUSSION

There have been a number of reports of spontaneous extrusion of the peritoneal end of the VP shunt, with per oral and per anal extrusions being the most common sites. Rarer sites are through the intact chest wall or abdominal wall or the umbilicus.[123]

The spontaneous extrusion of the distal end of the shunt is related to its ability to perforate any intra-abdominal hollow viscus and occasionally the anterior abdominal wall itself. The constant pressure of the tip of the catheter on the skin or the viscera causes local inflammation and a strong fibrotic reaction surrounding the distal end, ultimately eroding its surface. The perforation is usually asymptomatic, and spontaneous appearance of the catheter from an orifice brings attention to the problem.[4]

The earliest reported case of spontaneous umbilical fistula was in 1973 by Adoleye et al.[5] Mohindra et al. noted five possible causes of such a complication such as anatomical weakness, umbilical abscess, persistent umbilical vein, perivisceritis, and persistent urachal remnant while adding abnormally patent intercoelomic communication as a cause in their case.[6]

In our case, migration of the lower end of the shunt was seen near the umbilicus, with the formation of an umbilical fistula that was discharging CSF [Figures 1 and 2]. The surrounding pouting granulation tissue was probably formed as a result of secondary infection with subsequent healing by fibrosis that resembled an umbilical adenoma.

We theorized that peristaltic movements in the peritoneal cavity pushed the distal end of the shunt toward the umbilicus, and a persistent patent congenital opening like vitello-intestinal duct provided the pathway for the shunt to extrude through the umbilicus. The superadded infection was probably a result of the CSF fistula and not a cause.[5]

Further, spontaneous umbilical fistulas can have either an early or a delayed presentation. Early presentation within 6 months of shunt surgery is seen when there is shunt infection or wound infection as evidenced by the case of Antunes and Ribeiro in 1975 and Wani et al. in 2002.[78] Delayed presentation is usually related to congenital anomalies in the form of patent vitello-intestinal duct or urachus or extraembryonic coelom as seen in our case. Our presentation is also unique in that the complication presented 4.5 years after the initial shunt surgery. It is delayed as it usually only manifests once the CSF production exceeded the CSF absorption that may have happened following mild peritoneal irritation.

It is well established that following the removal of the malfunctioning shunt, the patient may be shunt independent and not require CSF diversion. While Birbilis et al. and Dagtekin et al. presented cases with extrusion of the peritoneal catheters through the intact abdominal wall which were treated with shunt revisions, our patient remained asymptomatic and neurologically intact following shunt removal.[23]

During subcutaneous tunneling, it is advisable to tunnel and bury the peritoneal catheter at a deeper level beneath the skin to avoid epidermal trauma which predisposes to subsequent focal skin infection while simultaneously providing protection against dermal erosion by shunt components.[9] Avoiding surgical site infection and using soft-tipped catheters also help reduce the chances of spontaneous extrusion.

CONCLUSION

Spontaneous umbilical fistulas can be a rare complication due to distal migration of the peritoneal end of the VP shunt. Various pathophysiological forces may be responsible for this complication, and different theories are proposed for early and late presentations of these fistulas. Management in each case needs to be individualized and shunt revision is always not necessary. Tunneling in a deeper plane and avoiding surgical site infection are two ways to prevent this complication.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.