Bowel perforation and anal extrusion of the distal portion of a VP shunt (AEVPS) is a rare mechanic complication of VP shunts. Wilson et al. described the first case reported in 1966 [81]. Since then, more than 100 cases of bowel perforation have been reported and the incidence of this complication is thought to be around 0,1–1 % [15, 19, 27, 49, 58, 63, 72, 81, 82]. In a retrospective review of their series in 2006, Vinchon and colleagues found 19 cases of bowel perforation due to VP shunt: only three of them developed anal extrusion of the catheter.

Pathophysiology and risk factors Pathogenesis and predisposing factors are not completely well understood. By definition, AEVPS involves a bowel perforation that has been produced through different mechanisms. Di Rocco suggested that bowel erosion results from inflammation caused by a preexisting shunt infection [21]. Interaction between mechanic trauma and inflammation following infection can lead to the bowel perforation [18, 19, 27, 43]. In the majority of the reported cases pathogenic agents suggest a peritoneal focus (e.g., Escherichia coli), but it is not uncommon to find organisms that are typically related with a perioperative contamination (S. aureus or S. epidermidis). Different authors suggest that some cases of bowel perforation can be linked to mechanisms of inflammation and rejection of an infected foreign body [18, 19, 34]. Intestinal developmental alterations can be a predisposing factor too: in a paper from Matsuoka et al. bowel perforation was related to a duplication of ileum terminalis that resulted in continuous irritation of a fixed point on the bowel’s surface and finally perforation of the sigmoid colon [49]. Encasing fibrosis around the tube has been reported; this fibrosis may have an anchoring effect on the catheter, resulting in pressure and decubitus ulceration, which can lead to a perforation [6, 49]. It has been also reported that the weakness of certain peritoneal areas and the umbilical end of the vitellointestinal duct and the processus vaginalis into the scrotum might remain patent and act as a facilitating mechanism for perforation [59, 81]. Formerly it was believed that spring-coiled catheters were more prone to produce visceral perforations, as this kind of catheter had been implicated in more than the 50 % of the cases, and in cases of gastric and peroral migrations. However, many reports have shown that softer and more flexible silicon catheters can also produce bowel and other viscus perforation [21, 27, 63, 72].

No evidence has been found that the peritoneal opening technique (laparotomy versus trochar) when the shunt is positioned could be a risk factor for bowel perforation. Some authors consider that laparotomy is a safer technique; on the other hand, many papers find that trochar technique is at least as safe as laparotomy and it is not a risk factor for bowel perforation [78]. Perforations due to the trochar are described, but usually they are an operative, acute complication, not related to a chronic inflammatory process. Serious vascular damage has been reported during trochar insertion: for this reason, we favor mini-open laparotomy in opposition to trochar-guided insertion.

Other potential risk factors for bowel perforation could be suspected silicone allergy or weakness in the bowel wall resulting from deficient innervation like in children with myelomeningocele or congenital hydrocephalus [4, 12, 34]. It is not clear whether the length of the peritoneal catheter has any implication in hollow viscus perforations.

Diagnosis AEVPS is a pathognomonic sign of bowel perforation (Fig. 13.2), but it is infrequent (15.7 %). The absence of peritoneal signs is common in cases of bowel perforation due to a VP shunt. Clinical peritonitis is observed in 15–25 % of the cases. As much as 48 % of the cases can develop meningitis and/or ventriculitis. Abdominal radiology can be used when the diagnosis is not obvious. Abdominal CT scan with contrast and ultrasonography may show local inflammation signs and a thickened abdominal wall, but this exploration may result negative. In some cases with a high index of suspicion, it is possible to perform a shuntogram, which consists in the instillation of a contrast medium into the lower portion of the shunt, to demonstrate bowel perforation. Colonoscopy could be used to identify the point of the perforation [27, 49]. Cranial CT scan or MRI must complete the follow-up diagnostic studies. CSF sampling must be obtained from the shunt reservoir, from the tip of the catheter exposed or after externalization of the peritoneal end.

Treatment Bowel perforation has to be managed as a surgical emergency. If the patient does not show any symptom apart from the AEVPS, a conservative management is possible. The shunt can be cut in the abdominal surface, externalizing the proximal end and eliminating the distal end by a trans-anal traction. If the patient presents with clinical (peritoneal and/or meningeal) symptoms, imaging studies are needed to assess the presence of active hydrocephalus and localize the site of the bowel perforation. In this case, laparotomy or laparoscopy should be done in order to repair the intestinal perforation and retire the peritoneal end of the shunt [49, 82]. External ventricular drainage and antibiotic therapy is started until proven that CSF is not contaminated or, in case of CSF infection, that the infection has been defeated. After two or three consecutive negative CSF cultures, it is possible to replace the entire shunt system. Usually, new shunting to the peritoneum is an option, but atrial or pleural diversion is preferred when peritoneal positioning of the catheter is not feasible (e.g., big laparotomy, adhesions, or recurrent perforation).

Prognosis AEVPS is a rare but potentially severe complication of VP shunts. Children with no symptoms of perforation or meningitis show a better evolution. If intestinal perforation leads to a chemical or infectious peritonitis, or the patient suffers meningitis or ventriculitis after Gram-negative infection, then prognosis is bad. The highest mortality rates are shown in patients with abdominal complications [82].

Bladder perforation by a peritoneal catheter is another rare complication of VPS [6, 15, 35, 53, 63]. This complication was first reported by Grosfeld in two patients, aged 3 months and 1 year, in 1974 [32]. Catheter removal, bladder repair, supra-pubic cystostomy, and antibiotic therapy resulted in recovery in each case. Around ten cases involving normal (nonaugmented) urinary bladder have been reported [6, 15, 35]. It has been also reported more commonly after abdominal repair of urinary bladder, during augmentative procedures for neurogenic bladder [53]. The location of the bladder, which lies in the extraperitoneal compartment, makes it a highly unlikely site of peritoneal catheter perforation, as the catheter must pass through the peritoneum into the extraperitoneal space and subsequently penetrate the bladder.

The mechanisms by which peritoneal catheters perforate hollow organs are not fully understood. An initial local inflammatory reaction around the tip of the catheter initiates its anchoring to the serosa of the hollow viscus [6, 19]. Calcifications occur at the distal tube in barium-impregnated catheters as sign of the fragmentation of the silicon polymers facilitating inflammation and anchoring [10]. Increase intra-abdominal pressure, peristaltic movements of the bowels, and CSF pulsation originates local pressure leading to necrosis and ultimate bladder perforation [6, 19, 63]. After entry into the urethra, final extrusion happens.

Shunt factors, which have been related to perforation, are sharp abdominal tip, long abdominal end of catheter, stiff consistency, barium coating of the catheter, or allergies to the shunt components (like silicon allergy). Initially, a higher tendency of perforation with spring-coiled catheters was reported. However, perforation may occur with any type of catheter.

Perforation of the bladder during trochar insertion or abdominal surgery can be prevented by draining the bladder before surgery. It is our policy to drain the bladder with a Foley catheter at the beginning of every surgery. If the patient is expected to have a long postoperative period (e.g., severe trauma or spinal cord tumor), Foley catheter is left in place. On the other hand, if shunting is a scheduled surgery, Foley is retired immediately after the procedure.

It has been reported that silicone rubber has a slight tendency to stick when it is in a dry state [35]. In distal independent catheters, the system may be flushed with saline. In unishunt systems, at least all components must be moisturized with saline and/or gentamycin dilution.

On examination, patients can be afebrile without neurological deficits or meningeal signs. Abdomen signs may be absent and peritoneal catheter can be seen coming out of the urethra with drops of CSF from the distal end. This can be explained because of a sealing effect of peritoneum around the catheter. If urine enters peritoneal cavity, patients may present with fever, abdominal pain, distension, and erythema of the abdominal wall between the umbilicus and the pubis. Diagnostic imaging must include again abdominal ultrasound and CT scanning.

Treatment In case that a bladder perforation occurs, it must be treated as an emergency. The shunt can be cut proximally and pulled out through the urethra. If no irritative signs are present or abdominal imaging excludes intraperitoneal complication (such as pneumoperitoneum or urinoma) a Foley catheter is placed and one may allow the bladder to heal on its own. However, at the lesser symptom of peritoneal defense, bladder can be approached extraperitoneally, repaired, and cystostomy or transient Foley catheter performed. In both cases, shunt is externalized proximally or external ventricular drainage left in place until CSF cultures are negative or peritoneum is ready to receive again the shunt in the opposite side.

Following a similar pathological pattern, perforation of the scrotum, umbilicus, and vagina has been also reported.

Other rare complications of VP shunt surgery include migration, intestinal volvulus, abdominal wall perforation, or VPS-related abdominal metastases originating from brain tumors. Acute cholecystitis after subphrenic suprahepatic abscess among other bizarre complications is reported occasionally in literature.

After multiple distal failures, the peritoneal cavity is often deemed unsuitable for cerebrospinal fluid (CSF) diversion. Probably more than 30 % of patients with VP shunts will experience abdominal complications [1, 9, 18, 22]. Different places have been described when peritoneum is considered inadequate or impaired for CSF diversion, including right atrium of the heart, pleural space, or gall bladder among others. However, because of the benefits of VP over VA shunting (mostly a lower rate of severe complications) [25, 51, 78], many authors defend that every effort should be made to preserve the peritoneum as the place for the definitive distal catheter [1, 23, 37, 41, 62]. Potential complications of ventriculoatrial shunt malfunction are thromboemboli and infection, both life threatening [51, 78]. Thromboembolism occurs in 0.3 % of patients and is nearly always fatal [51]. Shunt nephropathy is also a potential risk long time recognized and reported in the past. The treatment of occluded or sluggish catheters is often anticoagulation that exposes the patient to severe bleeding risks, and revision of these catheters, when needed, is related to more complex procedures that include manipulation of central veins. Revision rates due to catheter outgrowth have been reported to be as high as 66 % [78].

For all these reasons, every effort must be made to find a functional peritoneum after previous shunt failures due either to infection or distal occlusion. Laparoscopy has been used as in the management of VP shunt complications since the 1970s, providing excellent visualization of the abdomen through a minimally invasive approach [1, 37, 41, 42]. Surgeon can examine the abdomen for adhesions, cysts, loculations, and laparoscopy can be used to clear distal obstructions, excise CSF pseudocysts, or successfully reposition the distal catheter in patients with previous multiple shunt revisions or distorted abdominal anatomy. Laparoscopy is an excellent tool in locating suitable peritoneal pockets of the peritoneum where the distal catheter can be successfully replaced [1, 41, 42]. Open laparotomy remains an alternative for abdominal exploration, in more complex cases where laparoscopy is not indicated or easy to perform due to multiple loculations or dense adhesions [50, 62].

An alternative approach to minimize the effects of peritoneal adhesions is the employment of “reserve pouches” inside the peritoneal cavity [9, 23, 50, 57, 62]. Typically, retrohepatic subdiaphragmatic recess has been successfully used after “lost peritoneum” [57]. Several reports emphasize the advantages: it is away from previous adhesions, which usually affect the anterior peritoneum; and it is the place where the absorption rate of the peritoneal fluid through the lymphatics is higher, due to concentration of lymphatic ducts, and the presence of intracellular channels (“stomata”) at the diaphragmatic peritoneum. Rengachary described a transthoracic transdiaphragmatic VP shunt for patients with difficult access to the peritoneal cavity due to diffuse anterior peritoneal adhesion. In this approach, the catheter is inserted directly into the suprahepatic subdiaphragmatic space through the thoracic cavity [67].