Intraabdominal Complications Secondary to Ventriculoperitoneal Shunts: CT Findings and Review of the Literature

Gastrointestinal Imaging • Original Research Chung et al. CT of Ventriculoperitoneal Shunts Downloaded from www.ajronline.org by 37.44.207.67 on 01/2...
Author: Horatio Cox
4 downloads 2 Views 837KB Size
Gastrointestinal Imaging • Original Research Chung et al. CT of Ventriculoperitoneal Shunts

Downloaded from www.ajronline.org by 37.44.207.67 on 01/21/17 from IP address 37.44.207.67. Copyright ARRS. For personal use only; all rights reserved

Gastrointestinal Imaging Original Research

Intraabdominal Complications Secondary to Ventriculoperitoneal Shunts: CT Findings and Review of the Literature Jae-Joon Chung1 Jeong-Sik Yu1 Joo Hee Kim1 Se Jin Nam1 Myeong-Jin Kim 2 Chung JJ, Yu JS, Kim JH, Nam SJ, Kim MJ

Keywords: abdominopelvic CT, cerebrospinal fluid, hydrocephalus, peritonitis, pseudocyst, ventriculo­ peritoneal shunt DOI:10.2214/AJR.09.2463 Received January 27, 2009; accepted after revision April 12, 2009. Supported by a research grant from Yonsei University College of Medicine for 2008. 1 Department of Radiology and Research Institute of Radiological Science, Gangnam Severance Hospital, Yonsei University College of Medicine, 612, Eunjuro, Gangnam-gu, 135-720 Seoul, Korea. Address correspondence to J. J. Chung ([email protected]). 2 Department of Radiology and Research Institute of Radiological Science, Severance Hospital, Yonsei University College of Medicine, Seoul, Korea.

AJR 2009; 193:1311–1317 0361–803X/09/1935–1311 © American Roentgen Ray Society

AJR:193, November 2009

OBJECTIVE. The purpose of our study was to evaluate the abdominopelvic CT findings of various intraabdominal complications secondary to ventriculoperitoneal shunts for hydrocephalus and to review the literature. MATERIALS AND METHODS. The CT images of 70 patients (33 men and 37 women; mean age, 48.5 years) who underwent ventriculoperitoneal shunt placement and abdominopelvic CT because of shunt-related abdominal symptoms were reviewed retrospectively. CT images were analyzed with regard to the location of the shunting catheter tip; site, size, wall, and septa of localized fluid collection; peritoneal thickening; omentomesentery infiltration; abscess; bowel perforation; abdominal wall infiltration; and thickening of the catheter track wall. RESULTS. The mean period between the last ventriculoperitoneal shunting operation and CT was 11 months (range, 1 week to 115 months), and the mean number of ventriculoperitoneal shunting operations undergone was 1.4 (range, 1–6). A total of 76 ventriculoperitoneal shunting catheters were introduced in 70 patients: 64 patients had a unilateral catheter inserted and six patients had bilateral catheters inserted. Sixteen patients (22.9%) were pathologically diagnosed with ventriculoperitoneal shunt–related complications: 11 cases (15.7%) of shunt infection, six cases (8.6%) of CSF pseudocyst, four cases (5.7%) of abdominal abscess, three cases (4.3%) of infected fluid collection, and one case (1.4%) of bowel perforation. Microorganisms were cultured from the tip of the shunting catheter or peritoneal fluid in 11 patients (15.7%). CONCLUSION. On abdominopelvic CT, various intraabdominal complications secondary to ventriculoperitoneal shunt were shown, of which, shunt infection was the most common, followed by CSF pseudocyst, abscess, and infected fluid collection.

P

lacement of a ventriculoperitoneal shunt is the most common operation performed in the treatment of hydrocephalus. Intraabdominal complications after ventriculoperitoneal shunt placement are most commonly located near the peritoneal end of the shunt catheter; more than 50% of patients require shunt revision [1, 2]. The most common complications have been infection of the shunt, malfunction due to blockage, disconnection, migration, and equipment failure, which are related to extraperitoneal retraction of the catheter, development of an incisional hernia, subcutaneous collection of CSF, and peritoneal pseudocyst formation due to lowgrade infection followed by wrapping by the omentum [2–4]. Other complications reported in the literature include intestinal perforation, CSF ascites, inguinal hernia, and intestinal volvulus [5–7]. These complications

may manifest as either local abdominal signs or increased intracranial pressure. Shunt infection remains a frequent and potentially fatal complication of CSF diversion. Therefore, a key issue in the treatment of these complications is early and correct diagnosis of intraabdominal complications by CT, MRI, sonography, or abdominal radiography. There have been several case reports about various abdominal complications that can occur after ventriculoperitoneal shunting operations [8–11]. The purpose of this study was to evaluate the CT findings of various intraabdominal complications secondary to ventriculoperitoneal shunt placement for hydrocephalus and to review the literature. Materials and Methods This retrospective study was approved by our institutional review board, and the requirement for

1311

Downloaded from www.ajronline.org by 37.44.207.67 on 01/21/17 from IP address 37.44.207.67. Copyright ARRS. For personal use only; all rights reserved

Chung et al. informed consent was waived. We reviewed the medical records of all patients who underwent abdominopelvic CT after ventriculoperitoneal shunt placement from January 2001 to June 2008. The medical records of 100 consecutive patients were reviewed. An attending abdominal radiologist retrieved the images from a hospital archiving system and conducted a preliminary review of the CT images of the 100 patients. Among the patients reviewed, a total of 30 were excluded for the following reasons: seven patients had no specific abdominal symptoms (staging workup for underlying cancer [n = 4], evaluation of hypertension [n = 2], and falling injury [n = 1]); five patients had no recorded clinical information for an abdominopelvic CT scan; five patients underwent ventriculoperitoneal shunt removal just before the CT for shunt malfunction; four patients underwent CT for evaluation of percutaneous endoscopic gastrostomy tube placement; three patients underwent CT for renal or ureteral stones; three patients underwent CT for urinary tract infection; two patients underwent CT for gallstones; and one patient underwent CT for voiding difficulty. Inclusion criteria for the study included abdominopelvic CT with previous ventriculoperitoneal shunt placement and abdominal symptoms such as diffuse or localized abdominal pain, fever with abdominal symptoms, advanced hydrocephalus because of shunt malfunction, ventriculitis because of shunt infection, prolonged diarrhea, abdominal wall mass, pus discharge from the operation site, and hematochezia. After exclusion of the aforementioned 30 patients, a total

of 70 patients (33 male and 37 female; age range, 2–75 years; mean age, 48.5 years) were enrolled in this study. The causes of hydrocephalus in the 70 patients were 14 cases of primary or secondary brain tumors, 12 cases of subarachnoid hemorrhage and intraventricular hemorrhage (IVH), 11 cases of traumatic intracerebral hemorrhage (ICH) and IVH, six cases of postoperative ischemic change, six cases of postoperative hemorrhage, five cases of hypertensive ICH and IVH, four cases of IVH, three cases of tumor seeding, two cases of vascular ICH and IVH, two cases of ICH after infarct, two cases of congenital disease, two cases of meningitis, and one case of subdural hematoma. Two cases of vascular ICH and IVH were from a case of moyamoya disease and a case of arteriovenous malformation (AVM). Two cases of congenital diseases were from a case of congenital hydrocephalus and a case of ArnoldChiari malformation. The window for abdominopelvic CT was from 2 cm above the right hemidiaphragm to 2 cm below the symphysis pubis, with a 7-mm collimation and a 7-mm reconstruction interval. After IV administration of 120–150 mL of nonionic iodinated contrast medium (iopromide, Ultravist 300, Schering), given by an automatic injector at 3–4 mL/s, 3-minute delayed equilibrium phase images were obtained. All scanned images were sent to the PACS for interpretation. The locations of intraperitoneal fluid collections were divided into right upper quadrant (RUQ), right lower quadrant (RLQ), left upper quadrant (LUQ), left lower quadrant (LLQ), right paracolic

gutter (RPG), left paracolic gutter (LPG), midabdomen (paraumbilical area), pelvic cavity (below acetabular roof), and abdominal wall. One abdominal radiologist with 14 years of experience reviewed all the abdominopelvic CT images, focusing on the site and size of localized fluid collection, the presence of walls or septa in the localized fluid collections, peritoneal thickening, peritoneal contrast enhancement, omentomesentery infiltration, abscess, bowel perforation, bowel-wall thickening and contrast enhancement, abdominal wall infiltration, thickening of the catheter track wall, and intraperitoneal free air.

Results Abdominopelvic CT was performed in 70 patients for complaints of varying abdominal symptoms, such as diffuse abdominal pain (n = 25), localized abdominal pain (n = 17), aggravated hydrocephalus due to shunt malfunction (n = 9), fever with vague abdominal symptoms (n = 6), ventriculitis with suspected shunt infection (n = 5), prolonged diarrhea (n = 3), palpable abdominal mass (n = 2), subphrenic free air (n = 1), pus discharge from the operation site (n = 1), and hematochezia (n = 1). Twelve (70.6%) of the 17 patients with localized abdominal pain complained of lower abdominal pain. Among the 12 patients with lower abdominal pain, six had RLQ pain that was initially diagnosed as acute appendicitis, two (2.9% of all patients) of whom were finally diagnosed with acute appendicitis.

A

B

Fig. 1—51-year-old woman with large pseudocyst in abdominal wall. A, Contrast-enhanced abdominopelvic CT scan shows about 13 × 10 cm pseudocyst in subcutaneous layer of right anterior abdominal wall with internal ventriculoperitoneal shunt catheter. Smoothly compressed abdominal wall muscle is noted. Causes of hydrocephalus were subarachnoid hemorrhage and intraventricular hemorrhage. B, More caudal CT image shows pericystic irregular fluid collections (arrows) in abdominal wall. This patient complained of aggravated headache because of advanced hydrocephalus from shunt malfunction. Subsequently, shunt revision was performed, and there was no growth of microorganism from catheter tip.

1312

AJR:193, November 2009

Downloaded from www.ajronline.org by 37.44.207.67 on 01/21/17 from IP address 37.44.207.67. Copyright ARRS. For personal use only; all rights reserved

CT of Ventriculoperitoneal Shunts Ventriculitis was suspected in five patients because of findings on brain CT or MRI examinations that were performed for acute and persistent headache or increased intra­ cranial pressure. The mean period between the last ventriculoperitoneal shunting operation and abdominopelvic CT was 11 months (range, 1 week to 115 months), and the mean number of ventriculoperitoneal shunting operations undergone was 1.4 (range, 1–6). In the 70 patients, 76 ventriculoperitoneal shunting catheters were introduced: 64 patients with a unilateral catheter and six patients with bilateral catheters. There were 46 right ventricle–origin shunting catheters and 30 left ventricle–origin shunting catheters; six patients had bilateral ventriculoperitoneal shunting catheters placed. In the 70 patients, 87 localized fluid collection sites, of which 47 patients (67.1%) had one fluid collecting site, 17 (24.3%) had two fluid collecting sites, and two (2.9%) had three fluid collecting sites, as well as six CSF pseudocysts were found. The locations of the 87 localized fluid collection sites were the pelvic cavity (n = 55), RPG (n = 16), mid­ abdomen (n = 5), LPG (n = 4), RLQ (n = 3), abdominal wall (n = 1), RUQ (n = 1), LUQ (n = 1), and LLQ (n = 1). Among them, one patient had fluid collections in both paracolic gutters. The locations of the six CSF pseudocysts were the RLQ (n = 3), pelvis (n = 2), and abdominal wall (n = 1) (Fig. 1).

Fig. 2—33-year-old man with localized omentomesentery infiltration. Contrast-enhanced abdominopelvic CT scan shows localized dirty infiltration in omento­mesentery of abdominal left lower quadrant (LLQ) with adjacent peritoneal thickening and minimal fluid collection. Ventriculoperitoneal shunt catheter (arrows) is noted within localized infection area. Cause of hydrocephalus was postoperative intraventricular hemorrhage for arteriovenous malformation and patient complained of localized abdominal pain in abdominal LLQ. Shunt revision was performed and Staphylococcus aureus grew from tip of shunt catheter.

Of the 70 patients, 16 (22.9%) were histopathologically diagnosed with shunt-related complications, which included 11 cases (15.7%) of shunt infection (Fig. 2), six cases (8.6%) of CSF pseudocyst formation (Fig. 3), four cases (5.7%) of abdominal abscess (Fig. 4), three cases (4.3%) of infected localized fluid collection (Fig. 5), and one case (1.4%) of bowel perforation. Among these, all four cases of abscess, three cases of pseudocyst, and two cases of infected localized fluid collection coexisted in the patients with shunt infection. Of the six cases (8.6%) of CSF pseudocysts, five were located in the peritoneal space and one was located in the abdominal wall. One case (Fig. 6) of pseudocyst caused small-bowel obstruction secondary

A

to wrapped mesentery at the catheter tip within 5 days of ventriculoperitoneal shunt placement. Three cases of infected fluid collection were diagnosed by sonographically guided aspiration of fluid. One patient with subphrenic free air was diagnosed with peritonitis secondary to a 5-mm bowel perforation in the transverse colon due to the shunting catheter. Eighteen patients (25.7%) showed a wall of fluid collection, 12 (17.1%) of which showed contrast enhancement of the wall. Seven cases (10.0%) showed internal septa within the fluid collection. There were 30 cases (42.9%) of peritoneal thickening, 19 cases (27.1%) of peritoneal enhancement, 34 cases (48.6%) of omentomesentery infiltration (Figs. 2 and 5),

B

Fig. 3—20-year-old man with intraperitoneal pseudocyst and inflammatory infiltration in abdominal wall. A, Contrast-enhanced abdominopelvic CT scan shows 11 × 6 cm pseudocyst in abdominal right lower quadrant with enhanced wall and internal shunt catheters (arrows). Small amount of fluid collection is also seen in both paracolic gutters. Causes of hydrocephalus were traumatic hemorrhage and postoperative ischemic change. CT was performed because of abdominal pain. B, More cephalic CT image shows inflammatory fatty infiltration in subcutaneous layer of right midabdominal wall with adjacent skin thickening. Adjacent small-bowel loops show mild wall thickening with mesenteric haziness and fluid collections in both paracolic gutters. There was no growth of microorganisms from catheter tip and shunt externalization was performed.

AJR:193, November 2009

1313

Downloaded from www.ajronline.org by 37.44.207.67 on 01/21/17 from IP address 37.44.207.67. Copyright ARRS. For personal use only; all rights reserved

Chung et al. Fig. 4—14-year-old boy with intraperitoneal abscess. Contrastenhanced abdominopelvic CT scan shows 3 × 1 cm lentiform-shaped low-density lesion (arrows) with rimlike enhancement noted in left anterior peritoneal cavity with adjacent peritoneal thickening. Cause of hydrocephalus was congenital Arnold-Chiari malformation. CT was performed because of suspicion of shunt malfunction. Two shunt catheters are noted in both sides of abdominal wall with distal catheter near descending colon. This lesion was confirmed as abscess by percutaneous needle aspiration, and Pseudomonas aeruginosa grew from aspirated abscess fluid. Shunt catheter was removed.

four cases (5.7%) of abscess, one case (1.4%) of bowel perforation, nine cases (12.9%) of bowel-wall thickening, 38 cases (54.3%) of abdominal wall infiltration (Fig. 3), and seven cases (10.0%) of thickening of the catheter track wall in the abdominal wall. Specimen cultures from the peritoneal tips of the shunt catheter or intraperitoneal fluid were obtained in 21 patients (30.0%). The following microorganisms grew in 11 patients (15.7%): five cases of Staphylococcus aureus, one case of Pseudomonas aeruginosa, one case of vancomycin-resistant enterococci, one case of Acinetobacter baumannii, one case of Micrococcus species, one case of gram-positive cocci, and one case of gram-negative cocci. In 10 patients, there was no growth.

For the treatment of ventriculoperitoneal shunt–related abdominal complications in the 16 patients (22.9%), the following were used: antibiotic therapy only (n = 4), shunt externalization and antibiotic therapy (n = 2), exploratory laparotomy (n = 2), shunt revision (n = 2), sonographically guided percutaneous drain tube insertion (n = 2), removal of the catheter (n = 2), laparotomy and drainage (n = 1), and conservative treatment (n = 1). Discussion The use of the peritoneal cavity for CSF absorption in ventriculoperitoneal shunting was first introduced in 1908 by Kausch [12]. Other shunting techniques have since been used and include a ventriculoatrial shunt, a

A

lumboperitoneal shunt, and a third ventriculostomy [13]. The peritoneal cavity is preferable to the pleural cavity for insertion or reinsertion of the shunt [6]. The most common causes of shunt malfunction are catheter obstruction and infection. The incidence of ventriculoperitoneal shunt– related abdominal complications has been reported to be from 5% to 47% [14, 15]. In our study, 16 patients (22.9%) were histopathologically diagnosed with shunt-related complications. The most common distal ventriculoperitoneal shunt complications include shunt infection, subcutaneous collection of CSF, peritoneal pseudocyst, bowel perforation, intestinal volvulus [6], mesenteric pseudotumor, migration of the catheter into the pleural cavity and heart, and development of an incision hernia [8–11, 16–20]. Other less-common abdominal complications [21, 22] include bowel obstruction secondary to adhesion; subphrenic abscess, cerebrospinal–enteric fistula; untreatable CSF ascites; catheter disconnection; and extraperitoneal retraction of the catheter through the mouth [10], umbilicus [4], bladder, vagina, anus [11], or scrotum. Nonenteric viscus perforations also can occur and can involve multiple organs, such as the gallbladder, stomach, liver, uterus, or urethra. Obstruction of the distal catheter must be treated as an emergency because it can lead to a significant increase in intracranial pressure, resulting in associated complications that can cause considerable morbidity and possibly death [5].

B

Fig. 5—67-year-old woman with intraperitoneal infected fluid collection. A, Contrast-enhanced abdominopelvic CT scan shows large amount of ascites with thickened and enhanced peritoneum (arrows) with ventriculoperitoneal shunt catheter in left lower abdomen. Causes of hydrocephalus were subarachnoid hemorrhage and intraventricular hemorrhage. CT was performed because of abdominal pain, nausea, and vomiting. B, More cephalic CT image shows loculated fluid collections with thin wall in both sides of midabdomen. Dirty fatty infiltration is also seen in omentomesentery of right midabdomen, with fluid collections in right paracolic gutter and mesentery. Micrococcus species grew from tip of shunt catheter. External drainage tube was inserted for treatment.

1314

AJR:193, November 2009

Downloaded from www.ajronline.org by 37.44.207.67 on 01/21/17 from IP address 37.44.207.67. Copyright ARRS. For personal use only; all rights reserved

CT of Ventriculoperitoneal Shunts

A

B

Fig. 6—48-year-old man with intraperitoneal pseudocyst causing small-bowel obstruction. A, Contrast-enhanced abdominopelvic CT scan shows approximately 6 × 5 cm pseudocyst in abdominal right lower quadrant with adherent and distended neighboring small-bowel loops. Shunt catheter (arrow) is seen just below peritoneum. Cause of hydrocephalus in this patient was chondrosarcoma in skull base, and CT was performed because of abdominal pain. B, More cephalic CT image shows considerably more distended small-bowel loops with internal bowel contents and air–fluid level, suggesting mechanical bowel obstruction. Laparoscopy confirmed small-bowel obstruction caused by adherent pseudocyst secondary to wrapped mesentery at catheter tip. Ventriculoperitoneal shunt catheter was removed and reintroduced into another site. No microorganism grew from pseudocyst fluid.

Malfunction of the ventriculoperitoneal shunt after initial placement occurs in approximately 25–35% of patients at 1 year [23], and 70–80% of patients require at least one revision at some point in their lives [16]. Cochrane and Kestle [24] reported the initial shunt failure rate to be 31% at 6 months for experienced surgeons, with an infection rate of 7% over the same period. In our study, the mean period between the last ventriculoperitoneal shunting operation and abdominopelvic CT was 11 months; 52 patients (74.3%) had an interval of less than 11 months. Among these 52 patients, eight underwent abdominopelvic CT within 1 week because of abdominal symptoms. One of these eight patients underwent laparotomy because of small-bowel obstruction caused by smallbowel mesentery wrapping around the catheter tip and pseudocyst. Shunt infection remains a frequent and potentially fatal complication of CSF diversion, with a reported incidence of 5–47% [14, 15], and approximately 70% of shunt infections occur within 2 months after shunt placement [13, 25]. In our study, shunt infection was confirmed by bacterial culture in 11 patients (15.7%). The most common organism was S. aureus. However, it has been reported that 7% of ventriculoperitoneal shunt infections are caused by Escherichia coli [17]. Peritoneal fluid is either absent or present in only a small amount in patients with normally functioning ventriculoperitoneal

AJR:193, November 2009

shunts. CSF loculation may present as recurrent ascites, a peritoneal cyst, an omental cyst, or subphrenic or lesser sac loculation [8]. Peritoneal CSF pseudocyst formation is an unusual complication, with a reported incidence of less than 1.0–4.5% [8, 26]. The wall of the pseudocyst is composed of fibrous tissue or an inflamed serosal surface without an epithelial lining and is filled with CSF and debris [8, 27]. The most common presentation of an abdominal CSF pseudocyst in children is elevated intracranial pressure and abdominal pain, whereas local abdominal signs, such as abdominal pain, distention, nausea, or vomiting, predominate in adults [8]. An abdominal CSF pseudocyst was first described by Harsh [28] in 1954. Hahn et al. [29] reported that infection was the most prominent cause of pseudocyst formation (80%) and emphasized that all cases of abdominal pseudocysts should be considered to be caused by infection until proven otherwise. The most common intraabdominal response to infection is sheathing of the peritoneal catheter. The CSF draining into these sheaths may produce large intraabdominal fluid-filled cysts [8]. The infection and subsequent high levels of CSF protein, allergic reactions to immunization [30], liver dysfunction [19], and tissue reaction against tubing material and CSF protein [20] have been known to impair the absorption of CSF and to have a role in pseudocyst formation.

The time from the last shunting procedure to the development of an abdominal pseudocyst ranges from 3 weeks to 5 years [16]. There has been a reported case of pseudocyst formation 10 years after ventriculoperitoneal shunt placement. In our study, six cases (8.6%) of pseudocyst were detected, and the time interval between the last shunting operation and abdominopelvic CT ranged from 5 days to 25 months. The CSF pseudocyst can either move freely within the peritoneal cavity or adhere to loops of small bowel, the serosal surface of solid organs, or the parietal peritoneum [8]. CSF pseudocysts can be differentiated from ascites by their characteristic displacement of the bowel gas pattern on abdominal films and by the absence of shifting dullness [8]. Although sonography and CT can accurately localize abdominal fluid collections, differentiation of ascites from the aforementioned cystic lesions may not be possible. Therefore, fine-needle aspiration of the localized CSF collections under sonographic or CT guidance should be performed to increase the diagnostic yield. Coley et al. [31] reported that although the sonographically guided percutaneous aspiration of CSF pseudocyst was not curative, performance of this procedure to alleviate the acute symptoms followed by elective shunt revision is a feasible alternative to the traditional treatment approach and could be helpful to limit radiation exposure to patients who were likely

1315

Downloaded from www.ajronline.org by 37.44.207.67 on 01/21/17 from IP address 37.44.207.67. Copyright ARRS. For personal use only; all rights reserved

Chung et al. to have significant exposure during their lives, especially children. If infection is present, the pseudocyst wall should be excised and the peritoneal shunting catheter removed [8]. In our study, 18 cases (25.7%) showed a wall of fluid collection, of which 12 cases (17.1%) showed wall enhancement and seven (10.0%) showed internal septa; however, histopathologic diagnoses via fluid aspiration were available in only a few cases. Once the shunt tip is removed, the pseudocyst gradually collapses because there is no secretory epithelium present in the cyst [3]. The formation of a CSF pseudocyst is a poor prognostic sign for the usefulness of the peritoneal cavity for shunting [32]. Although previous abdominal pseudocyst formation and peritonitis are not contraindications to subsequent peritoneal shunting in some reports [19, 29], the CSF had to be diverted to other cavities because of either recurrence of the cysts or failure of the peritoneum to absorb fluid [27]. Culture of the tip of the peritoneal catheter was reported to be more sensitive than culture of the CSF [8]. Bowel perforation is a rare complication of ventriculoperitoneal shunt placement, occurring in less than 0.1–0.7% of cases [9], among which nearly half are diagnosed after removal of the catheter. Possible causes of bowel perforation include the sharp tip of the shunting catheter, subclinical shunt infection, and increased protein content in the CSF [11]. Recently, silicon allergy, which may result in a foreign body–like reaction, has been implicated in the breakdown and perforation of the bowel wall [18]. This complication can lead to fatal meningeal infection when not recognized early. The overall mortality rate of bowel perforation is nearly 15% in shunted patients [26]. In our study, one case (1.4%) of bowel perforation initially presented with subphrenic free air, and a subsequent abdominopelvic CT and exploratory laparotomy were performed, showing transverse colon perforation by the catheter tip. Patients with myelome­ ningocele or congenital hydrocephalus may be more prone to bowel perforation because of neurogenic weakness of the bowel wall from deficient innervations [7]. Bowel perforation can occur immediately after shunt placement or months or years later. Clinicians must be vigilant in their assessment for the presence of the following clinical conditions in shunted patients: meningitis or ventriculitis caused by an enteric microorganism (e.g., E. coli) and abdominal

1316

symptoms, such as prolonged unexplained diarrhea and fever [10]. Bowel perforation should be suspected in cases of shunt infection by gram-negative bacilli, pneumocephalus, prolonged unexpected diarrhea with sterile cultures, or abdominal pain [22]. The risk of ventriculoperitoneal shunt– related complications varies with the use of prophylactic antibiotics, size and condition of the patient, and experience of the operating surgeon. Recognition of these facts and consequent changes in treatment have resulted in a steady improvement in the outcomes of patients undergoing ventriculoperitoneal shunt placement [33]. The imaging techniques for early detection of intraabdominal complications secondary to ventriculoperitoneal shunt include radiography, sonography, CT, and MRI. On radiography, the location of the shunt tip, displaced bowel gases, soft-tissue mass of pseudocyst, and intraperitoneal free air can be seen. On sonography, the internal content, septa, and wall thickness of pseudocysts can be observed well. On contrast-enhanced CT or MRI, the peritoneal thickening, bowel-wall thickening and contrast enhancement, omentomesentery infiltration, abscess, axial location of shunt tip and adjacent abnormal findings, and localized extraluminal air densities because of bowel perforation can be well evaluated. Among these techniques, however, CT may be more useful for the exact diagnosis of complicated intraabdominal abnormalities. There are a few limitations to this study. The study is retrospective; therefore, 30 cases were excluded for various reasons. The number of pathologically confirmed cases was small, with only 15.7% of microorganisms being cultured. Infiltration or infection of the abdominal wall, catheter track, or omentomesentery was diagnosed only according to the findings on CT images of fatty infiltration or a thick and enhanced catheter track wall and were not proven histologically. The localized peritoneal fluid collections with contrast-enhanced walls or internal septa were not analyzed for possible infection or localized peritonitis. Nevertheless, we believe that familiarity with the broad spectrum of ventriculoperitoneal shunt complications will enhance the role of radiologists in the management of intraabdominal complications [7]. A high degree of suspicion and careful clinical and radiologic examinations could help diagnose and treat ventriculoperitoneal shunt–related complications before they progress to more serious conditions.

References 1. Borgbjerg BM, Gjerris F, Albeck MJ, Hauerberg J, Borgesen SE. Frequency and causes of shunt revisions in different cerebrospinal fluid shunt types. Acta Neurochir (Wien) 1995; 136:189–194 2. Sells CJ, Shurtleff DB, Loeser JD. Gram-negative CSF shunt associated infections. Pediatrics 1977; 59:614–618 3. Acharya R, Ramachandran CS, Singh S. Laparoscopic management of abdominal complications in ventriculoperitoneal shunt surgery. J Laparoendosc Adv Surg Tech 2001; 11:167–170 4. Wani AA, Ramzan A, Wani MA. Protrusion of a peritoneal catheter through the umbilicus: an unusual complication of a ventriculoperitoneal shunt. Pediatr Surg Int 2002; 18:171–172 5. Nfonsam V, Chand B, Rosenblatt S, Turner R, Luciano M. Laparoscopic management of distal ventriculoperitoneal shunt complications. Surg Endosc 2008; 22:1866–1870 6. Bal RK, Singh P, Harjai MM. Intestinal volvulus: a rare complication of ventriculoperitoneal shunt. Pediatr Surg Int 1999; 15:577–578 7. Sathyanarayana S, Wylen EL, Baskaya MK, Nanda A. Spontaneous bowel perforation after ventriculoperitoneal shunt surgery: case report and a review of 45 cases. Surg Neurol 2000; 54:388–396 8. Sharma AK, Pandey AK, Diyora BD, Mamidanna R, Sayal PP, Ingale HA. Abdominal CSF pseudocyst in a patient with ventriculo-peritoneal shunt. Indian J Surg 2004; 66:360–363 9. Martinez Hernández-Magro P, Román CB, Sáenz EV, Zavala MJ. Colonic perforation as a complication of ventriculoperitoneal shunt: a case report. Tech Coloproctol 2006; 10:353–355 10. Berhouma M, Messerer M, Houisssa S, Khaldi M. Transoral protrusion of a peritoneal catheter: a rare complication of ventriculoperitoneal shunt. Pediatr Neurosurg 2008; 44:169–171 11. Ghritlaharey RK, Budhwani KS, Shrivastava DK, et al. Trans-anal protrusion of ventriculo-peritoneal shunt catheter with silent bowel perforation: report of ten cases in children. Pediatr Surg Int 2007; 23:575–580 12. Kausch W. Die Behandlung des Hydrocephalus der kleinen Kinder. Arch Kiln Chir 1908; 87:709–796 13. Gupta DK, Dave S. Hydrocephalus. In: Gupta DK, ed. Textbook of neonatal surgery. New Delhi, India: Modern Publishers, 2000:434–450 14. Esposito C, Porreca A, Gangemi M, et al. The use of laparoscopy in the diagnosis and treatment of abdominal complications of ventriculoperitoneal shunts in children. Pediatr Surg Int 1998; 13:352–354 15. Lortat-Jacob S, Pierre-Kahn A, Renier D, et al. Abdominal complications of ventriculoperitoneal shunts in children: 65 cases. Chir Pediatr 1984; 25:17–21 16. Blount JP, Campbell JA, Haines SJ. Complications

AJR:193, November 2009

Downloaded from www.ajronline.org by 37.44.207.67 on 01/21/17 from IP address 37.44.207.67. Copyright ARRS. For personal use only; all rights reserved

CT of Ventriculoperitoneal Shunts in ventricular cerebrospinal fluid shunting. Neurosurg Clin N Am 1993; 4:633–656 17. Walters BC, Hoffman HJ, Hendrick EB, Humphreys RP. Cerebrospinal shunt infection: influences on initial management and subsequent outcome. J Neurosurg 1984; 60:1014–1021 18. Brownlee JD, Brodley JS, Schaefer IK. Colonic perforation by ventriculoperitoneal shunt tubing: a case of suspected silicon allergy. Surg Neurol 1998; 49:21–24 19. Latchaw JP Jr, Hahn JF. Intraperitoneal pseudocyst associated with peritoneal shunt. Neurosurgery 1981; 8:469–472 20. Rainov N, Schobess A, Heidecke V, Burkert W. Abdominal CSF pseudocyst in patients with ventriculo-peritoneal shunts: report of fourteen cases and review of literature. Acta Neurochir (Wien) 1994; 127:73–78 21. Paddon AJ, Horton D. Knotting of distal ventriculoperitoneal shunt tubing. Clin Radiol 2000; 55:1–3 22. Agha FP, Amendola MA, Shirazi KK, Amendola

BE, Chandler WF. Unusual abdominal complications of ventriculo-peritoneal shunts. Radiology 1983; 146:323–326 23. Vinchon M, Fichten A, Delestret I, Dhellemmes P. Shunt revision for asymptomatic failure: surgical and clinical results. Neurosurgery 2003; 52: 347–353 24. Cochrane DD, Kestle JR. The influence of surgical operative experience on the duration of first ventriculoperitoneal shunt function and infection. Pediatr Neurosurg 2003; 38:295–301 25. Kumar R, Singh V, Kumar MV. Shunt revision in hydrocephalus. Indian J Pediatr 2005; 72:843–847 26. Oh A, Wildbrett P, Golub R, Yu LM, Goodrich J, Lee T. Laparoscopic repositioning of a ventriculoperitoneal catheter tip for a sterile abdominal cerebrospinal fluid (CSF) pseudocyst. Surg Endosc 2001; 15:518 27. Parry SW, Schuhmacher JF, Llewellyn RC. Abdominal pseudocysts and ascites formation after ventriculoperitoneal shunt procedures: report of four cases. J Neurosurg 1975; 43:476–480

28. Harsh GR. Peritoneal shunt for hydrocephalus utilizing the fimbria of the fallopian tube for entrance to the peritoneal cavity. J Neurosurg 1954; 11: 284–294 29. Hahn YS, Engelhard H, McLone DG. Abdominal CSF pseudocyst: clinical features and surgical management. Pediatr Neurosci 1985–1986; 12: 75–79 30. Dean DF, Keller IB. Cerebrospinal fluid ascites: a comparison of a ventriculoperitoneal shunt. J Neurol Neurosurg Psychiatry 1972; 35:474–476 31. Coley BD, Shiels WE II, Elton S, Murakami JW, Hogan MJ. Sonographically guided aspiration of cerebrospinal fluid pseudocysts in children and adolescents. AJR 2004; 183:1507–1510 32. Fischer EG, Shillito J Jr. Large abdominal cysts: a comparison of peritoneal shunts: report of three cases. J Neurosurg 1969; 31:441–444 33. Pittman T, Williams D, Weber TR, Steinhardt G, Tracy T Jr. The risk of abdominal operations in children with ventriculoperitoneal shunts. J Pediatr Surg 1992; 27:1051–1053

F O R YO U R I N F O R M AT I O N

Mark your calendar for the following ARRS annual meetings: May 2–7, 2010—Manchester Grand Hyatt San Diego, San Diego, CA May 1–6, 2011—Hyatt Regency Chicago, Chicago, IL April 29–May 4, 2012—Vancouver Convention Center, Vancouver, BC, Canada April 14–April 19, 2013—Marriott Wardman Park Hotel, Washington, DC

AJR:193, November 2009

1317

Suggest Documents