Imaging of the Ventriculoperitoneal Shunt-Related Complications in Infants and Children With Hydrocephalus

PURPOSE: To evaluate the frequency and imaging findings of various ventriculo-peritoneal shunt-relatedcomplications in pediatric patients with hydrocephalus. MATERIALS AND METHODS: We retrospectively reviewed 246plain radiographs, three shuntograms, 53 ultrasounds, 133 CT scans, and 24 MR images obtained before and after theventriculo-peritoneal shunt procedure in 33 pediatric patients with hydrocephalus. Using preoperative images, theetiology of the hydrocephalus was assessed. Changes in the size and shape of the ventricles, the location andcontinuity of shunt apparatus, and the presence of any abnormal enhancement, hemorrhage, edema or tissue loss, orother findings of complications, were analyzed on postoperative images; the frequency and imaging findings ofshunt-related complications such as shunt malfunction, infection, hemorrhage or isolated ventricle, andcomplications caused by overdrainage, were thus evaluated. The frequency of such complications was analyzedaccording to the etiology of the hydrocephalus, and in addition, medical records were reviewed and correlated withimaging findings. RESULTS: In 18 of the 33 patients(54%), a total of 31 complications was detected. These werepresent in four of five cases (80%) of hydrocephalus caused by meningitis and ventriculitis, seven of twelve (58%)intraventricular hemorrhage, two of four (50%) unknown cases, three of nine (33%) congenital malformations, one oftwo (50%) tumors, and one (100%) congenital infection. Shunt malfunction was most common(n=15), and wasaccompanied by findings of enlarged ventricles, periventricular and peritubal edema, and abnormal location of theshunt tube. Symptoms and signs of increased intracranial pressure were also noted. Subdural hemorrhage andinfection were present infour cases each;findings of infection were enhancement of the ventricular wall, meninges,and parenchyma, as well as sonographically noted intraventricular septation and increased ventricular wall echo.Isolated lateral ventricle (n=4) or 4th ventricle (n=1) was persistently distended in spite of the presence of ashunt tube within the other ventricle. Collapsed ventricles, thick calvarium, and symptoms of increasedintracranial pressure were helpful in the diagnosis of slit ventricle syndrome(n=1). Immediate postoperativeintraventricular and parenchymal hemorrhage were noted in one case each. CONCLUSION: Shunt-related complicationswere found on postoperative images in 54% of pediatric patients with hydrocephalus; the frequency was highest incases of hydrocephalus caused by infection. Shunt malfunction occurred most frequently and each complication hadvarious imaging findings. For the diagnosis of post-operative shunt-related complication, continuous follow-upstudies are therefore necessary.

Imaging of the Ventriculoperitoneal Shunt-Related Complications in Infants and Children With Hydrocephalus

PURPOSE: To evaluate the frequency and imaging findings of various ventriculo-peritoneal shunt-relatedcomplications in pediatric patients with hydrocephalus. MATERIALS AND METHODS: We retrospectively reviewed 246plain radiographs, three shuntograms, 53 ultrasounds, 133 CT scans, and 24 MR images obtained before and after theventriculo-peritoneal shunt procedure in 33 pediatric patients with hydrocephalus. Using preoperative images, theetiology of the hydrocephalus was assessed. Changes in the size and shape of the ventricles, the location andcontinuity of shunt apparatus, and the presence of any abnormal enhancement, hemorrhage, edema or tissue loss, orother findings of complications, were analyzed on postoperative images; the frequency and imaging findings ofshunt-related complications such as shunt malfunction, infection, hemorrhage or isolated ventricle, andcomplications caused by overdrainage, were thus evaluated. The frequency of such complications was analyzedaccording to the etiology of the hydrocephalus, and in addition, medical records were reviewed and correlated withimaging findings. RESULTS: In 18 of the 33 patients(54%), a total of 31 complications was detected. These werepresent in four of five cases (80%) of hydrocephalus caused by meningitis and ventriculitis, seven of twelve (58%)intraventricular hemorrhage, two of four (50%) unknown cases, three of nine (33%) congenital malformations, one oftwo (50%) tumors, and one (100%) congenital infection. Shunt malfunction was most common(n=15), and wasaccompanied by findings of enlarged ventricles, periventricular and peritubal edema, and abnormal location of theshunt tube. Symptoms and signs of increased intracranial pressure were also noted. Subdural hemorrhage andinfection were present infour cases each;findings of infection were enhancement of the ventricular wall, meninges,and parenchyma, as well as sonographically noted intraventricular septation and increased ventricular wall echo.Isolated lateral ventricle (n=4) or 4th ventricle (n=1) was persistently distended in spite of the presence of ashunt tube within the other ventricle. Collapsed ventricles, thick calvarium, and symptoms of increasedintracranial pressure were helpful in the diagnosis of slit ventricle syndrome(n=1). Immediate postoperativeintraventricular and parenchymal hemorrhage were noted in one case each. CONCLUSION: Shunt-related complicationswere found on postoperative images in 54% of pediatric patients with hydrocephalus; the frequency was highest incases of hydrocephalus caused by infection. Shunt malfunction occurred most frequently and each complication hadvarious imaging findings. For the diagnosis of post-operative shunt-related complication, continuous follow-upstudies are therefore necessary.

Cine MR CSF Flow Study in Hydrocephalus: What are the Valuable Parameters?

PURPOSE: To evaluate the changes in diagnosis of intracranial cerebrospinal fluid (CSF) dynamics in the hydrocephalus, we studied the various parameters of cine phase contrast (PC) magnetic resonance (MR) CSF flow images in cases of acutely progressive hydrocephalus, comparing them with those in normal CSF circulation. MATERIALS AND METHODS: The MR images were obtained with a 1.5T (GE Signa, GE Medical Systems, Milwaukee, USA) unit using the 2 dimensional cine PC sequence with cardiac gating and gradient recalled echo imaging in ten cases of non-obstructive hydrocephalus (NOH), three of obstructive hydrocephalus (OH), and ten controls. The time-echo time-flip angle employed were 50 to 80 msec - 11 to 15 msec - 12 to 15 degrees. Temporal velocity information relating to cervical pericord CSF spaces, third and fourth ventricles, and the aqueduct were plotted as wave forms which were then analyzed for configuration, amplitude parameters (Vmax, Vmin, Vdif), and temporal parameters (R-S, R-SMV, R-D, R-DMV). The statistical significance of each parameter was examined using the paired t-test. All patients with OH underwent endoscopic third ventriculostomy, whereas all with NOH underwent shunting procedures. RESULTS: In five ROIs, distinct reproducible configuration features were obtained at aqueduct and cervical pericord spaces, but not at ventricular levels. We determined the statistically significant differences between controls and hydrocephalus patients using temporal, rather than amplitude parameters. In NOH, the graph showed R-DMV shortening (p<0.01) at the anterior cervical pericord space. In OH, there were R-DMV shortening (p<0.05) was seen at the anterior cervical pericord space, and R-DMV shortening (p<0.02) at the posterior cervical pericord space. In one case of OH, a typical change of configuration, mirror image, was obtained at aqueduct level, and in all OH cases, the level of obstruction could be determined. CONCLUSION: The results of cine PC MR CSF flow study may be valuable for points for determining the level of obstruction, explaining the cause of hydrocephalus, diagnosing OH, and deciding the need for shunting procedures; decisions would be based on an analysis of in vivo images, curve configuration, and phase shift of the velocity curve. In the diagnosis of acutely progressive hydrocephalus, termporal parameters are nore important than those relating to amplitude. Among the former, R-DMV and/or R-SMV may be the most valuable.

Intracranial Tuberculosis in Children: CT Findings before and after Treatment

OBJECT: To analyze the CT findings of intracranial tuberculosis in children at initial stage and during follow-up after treatment. MATERIALS AND METHODS: We evaluated 25 patients who were diagnosed by CSF analysis or response to anti-tuberculous medication as suffering from intracranial tuberculosis. There were 13 boys and 12 girls aged between 4 months and 14 years. Twenty-five initial and sixty-three follow-up CT scans were retrospectively analyzed. We evaluated the pattern of cisternal enhancement, the locations of infarction, and the presence of calcification and parenchymal granuloma. The changes of hydrocephalus and related complications, aswell as cisternal abnormality during anti-tuberculous medication were also evaluated. RESULTS: The initial findings on CT scan were hydrocephalus(75%), cisternal obliteration in precontrast study(64%), thick-line orring-shaped cisternal enhancement on postcontrast study(44%), infarctions(32%), calcifications(32%), periventricular edema(28%), and parenchymal granulomas(16%). On follow-up CT scan, hydrocephalus and cisternal enhancement had decreased to 35% and 82%, respectively, and the granulomas had changed to calcified nodules(100%). Ventriculo-peritoneal shunt or external ventricular drainage was needed in nine patients, and ventriculitis or complication of shunt procedure developed in four. CONCLUSION: Intracranial tuberculosis in children presented predominantly as meningitis involving basal cisterns and was associated with hydrocephalus. Infarction and calcification may be seen as parenchymal lesion. In spite of medical treatment, drainage was needed in about half the patients. During this treatment, the resolution of hydrocephalus, decreased cisternal enhancement, and calcification of the granulomas were seen.