HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Fujikawa, A. Articles by Nitatori, T. Search for Related Content PubMed PubMed Citation Articles by Fujikawa, A. Articles by Nitatori, T. Social Bookmarking                      What's this?

Comparison of MRI Sequences to Detect Ventriculitis

¹ Department of Radiology, Japan Self-Defense Forces Central Hospital, 1-2-24, Ikejiri, Setagaya, Tokyo 154-8532, Japan.
² Department of Radiology, Kyorin University Hospital School of Medicine, Tokyo, Japan.

Received December 18, 2004; accepted after revision August 22, 2005.

 
Address correspondence to A. Fujikawa.

Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
OBJECTIVE. The ability of different MRI sequences to depict characteristic findings suggestive of ventriculitis was compared.

CONCLUSION. The study comprised 20 brain MRI studies in 13 patients who had a final diagnosis of ventriculitis. Both diffusion-weighted imaging and FLAIR imaging were equally and highly sensitive for detecting intraventricular debris and pus—the most common MRI finding suggestive of ventriculitis. FLAIR imaging was superior to contrast-enhanced T1-weighted imaging for depicting ventricular wall abnormalities—a less common finding that also is suggestive of ventriculitis.

Keywords: brain • diffusion-weighted imaging • FLAIR imaging • infectious diseases • MRI

Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Ventriculitis is an uncommon CNS infection that has been described using a variety of terms including ependymitis, intraventricular abscess, ventricular empyema, and pyocephalus [1]. This variety of terms reflects various facets of the disease's pathologic process. Because ventriculitis is a severe intracranial infection that can lead to serious sequelae and death, prompt diagnosis is necessary. However, the clinical features of ventriculitis are often obscure and nonspecific. MRI plays an important role as a first-line diagnostic tool in the diagnosis of ventriculitis [2].

Previously reported characteristic MRI findings of ventriculitis include intraventricular debris and pus, abnormal periventricular and subependymal signal intensity, and enhancement of the ventricular lining on conventional MRI sequences [3, 4]. In addition, a few reports have illustrated the usefulness of diffusion-weighted imaging for detecting intraventricular debris and pus [1, 3, 5]. We have encountered cases of ventriculitis in which the above MRI features were subtle on some MR pulse sequences but obvious on others. Thus, knowing which MRI sequences are useful for detecting the characteristic findings of ventriculitis is important for daily clinical practice.

To the best of our knowledge, the utility of different MRI sequences for detecting the characteristic findings of ventriculitis has not been compared. In this study, we sought to determine which combinations of MRI sequences and findings were useful for the diagnosis of ventriculitis.

Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
By searching the computer database of our institution's department of radiology, we identified 13 patients who had been treated between November 1999 and July 2004 and were strongly suspected of having ventriculitis based on their MRI findings. The medical records of these patients were reviewed by two of the authors to obtain clinical follow-up information including laboratory, treatment, and outcome data and to confirm the final diagnosis based on the radiologic and clinical findings. The diagnoses of the 13 patients were consistent with ventriculitis. The patients, eight males and five females, had a mean age of 50.5 years (age range, newborn-85 years). The pathogens detected were Klebsiella pneumoniae (n = 1), Staphylococcus aureus (n = 1), Streptococcus pneumoniae (n = 2), Enterobacter cloacae (n = 1), Enterobacter aerogenes (n = 1), Escherichia coli (n = 1), Mycobacterium tuberculosis (n = 1), Pseudomonas aeruginosa (n = 2), Cryptococcus neoformans (n = 1), and unknown (n = 2). The specimens for cultures and Gram stains were obtained by lumbar puncture (n = 9), ventriculostomy tube (n = 3), and abscess drainage (n = 1). Three patients were clinically improved. Five patients had prolonged disturbance of consciousness. Five patients died during the study's time course (2-73 days after MRI). Underlying conditions of these patients included myelodysplastic syndrome (n = 1), lymphocytic leukemia (n = 1), postclipping of cerebral aneurysm (n = 1), malignant lymphoma (n = 1), pneumonia (n = 1), severe facial bone fracture (n = 1), AIDS (n = 2), mastoiditis and petrositis (n = 1), postresection of meningioma (n = 1), and cerebellar abscess (n = 1). Two patients had no risk factor. A total of 20 brain MRI studies performed in these 13 patients were included in the imaging analyses.

All MRI was performed using one of three 1.5-T imaging systems (Visart or Exelart, Toshiba Medical Systems; Gyroscan Intera, Philips Medical Systems). All MRI studies were performed according to our institution's protocol for brain infection and inflammation, although the imaging parameters varied slightly depending on the manufacturer of the system. The MRI protocol included axial fast spin-echo T2-weighted sequences (TR/TE range, 4,000-4,900/90-120; field of view [FOV], 18-22 x 22 cm; matrix, 160-192 x 256-384; number of excitations [NEX], 1-2; echo-train length, 13, 15, or 17; receiver bandwidth, 39.68, 41.856, or 62.464 kHz); axial contrast-enhanced spin-echo T1-weighted sequences (450-540/10-15; FOV, 18-22 x 22 cm; matrix, 160-176 x 258-384; NEX, 1-2); axial FLAIR sequences (8,000-10,000/105-120; FOV, 18 x 22; matrix, 160-192 x 256-320; NEX, 1-2; inversion time, 2,300-2,600 msec); and axial single-shot spin-echo echo-planar diffusion-weighted sequences with b values of 0 and 1,000 s/mm² along all three orthogonal axes (4,000-8,000/95-120; FOV, 22-25 x 26-30 cm; matrix, 128 x 128; NEX, 1). Calculated apparent diffusion coefficient (ADC) maps were also obtained.

Three radiologists with experience in brain imaging retrospectively reviewed a total of 100 hard-copy MR images. Hard-copy images of each sequence were separated from each other and randomly mixed. Each radiologist independently assessed the following findings: the presence of abnormal intraventricular signal intensity, the presence of abnormal periventricular signal intensity, and the presence of contrast enhancement of the ventricular wall on contrast-enhanced T1-weighted images. The presence of other MRI findings associated with intracranial inflammation, including hydrocephalus, meningitis, brain abscess, cerebritis, and choroid plexitis, was also assessed. Abnormal intensities on the diffusion-weighted images were correlated with corresponding findings on the ADC maps. The ADC values of abnormal intraventricular intensities and normal-appearing intraventricular CSF were obtained by calculating the mean values of three circular regions of interest (ROIs), each of which was 2-5 mm in diameter, on the ADC maps. For abnormal periventricular intensities on FLAIR images, each reviewer attempted to subjectively distinguish between the presence of an inflammatory process and an age-related normal variant. Interpretation discrepancies were resolved by the judgment of a third radiologist.

We followed our institution's ethical guidelines. At our institution, institutional review board approval and informed consent are not required for retrospective reviews of imaging studies and medical records.

Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Of the 20 MRI studies, intraventricular abnormal intensities were present in 19 (95%) of the diffusion-weighted images, 19 (95%) of the FLAIR images, 13 (65%) of the T2-weighted images, and 10 (50%) of the contrast-enhanced T1-weighted images. Of all MRI studies, abnormal intensities were detected in bilateral ventricles in 12 of the diffusion-weighted images, nine of the FLAIR images, five of the T2-weighted images, and five of the contrast-enhanced T1-weighted images. Abnormal intraventricular signal intensities relative to the signal intensities for normal CSF were hyperintense on the diffusion-weighted and FLAIR images, slightly hypointense on the T2-weighted images, and slightly hyperintense on the contrast-enhanced T1-weighted images (Figs. 1A, 1B, 1C, 1D, and 1E).

View larger version (143K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 1A —68-year-old man with severe mastoiditis and acute petrositis. T2-weighted image shows areas of slight hypointensity relative to CSF in bilateral trigone of lateral ventricle. This finding is suggestive of intraventricular debris and pus. Slight ventricular wall abnormality is noted.

View larger version (153K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 1B —68-year-old man with severe mastoiditis and acute petrositis. FLAIR image shows hyperintense intraventricular lesions relative to CSF and hyperintensity along ventricular lining, suggesting ependymitis.

View larger version (136K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 1C —68-year-old man with severe mastoiditis and acute petrositis. Contrast-enhanced T1-weighted image shows abnormal curvilinear enhancement along the ventricular wall. Intraventricular debris and pus are slightly hyperintense relative to CSF.

View larger version (84K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 1D —68-year-old man with severe mastoiditis and acute petrositis. Diffusion-weighted image shows areas of conspicuous intraventricular and periventricular hyperintensity, indicating restricted water diffusion in those areas.

View larger version (114K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 1E —68-year-old man with severe mastoiditis and acute petrositis. Apparent diffusion coefficient (ADC) map shows areas of decreased ADC values in corresponding lesions on diffusion-weighted image (D).

Abnormal periventricular intensities were detected in 17 (85%) of the FLAIR images, 11 (55%) of the diffusion-weighted images, and six (30%) of the T2-weighted images. Asymmetric periventricular abnormalities were seen in 12 of the FLAIR images, 10 of the diffusion-weighted images, and five of the T2-weighted images. Enhancement of the ventricular lining was observed in 12 (60%) of the 20 MRI studies. Asymmetric contrast enhancement of the periventricular region was seen in 10 studies. In 11 (55%) of the 20 MRI studies, hydrocephalus was observed in each of the sequences.

Specific MRI findings were found in association with a variety of conditions, including meningitis (n = 13), cerebral abscess (n = 2), cerebritis (n = 2), subdural empyema (n = 3), cerebellar abscess (n = 1), maxillary sinusitis (n = 2), mastoiditis with otitis media (n = 1), brain contusion (n = 2), postoperative changes (n = 3), and suspected choroid plexitis (n = 5).

Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
In our series, the most frequent sign of ventriculitis (95% both on diffusion-weighted and FLAIR images) was intraventricular debris and pus. Abnormal periventricular intensities or enhancements were observed less frequently (85% on FLAIR images and 60% on contrast-enhanced T1-weighted images). These results are in agreement with those of a previous study by Fukui et al. [3]. The present study showed that diffusion-weighted and FLAIR imaging concordantly showed intraventricular debris and pus in the same patients with an equally high detection rate. Thus, both of these sequences can be expected to contribute to the diagnosis of ventriculitis. However, our impression is that diffusion-weighted imaging is highly sensitive for detecting ventriculitis.

The number of lesions detected in bilateral ventricles was higher on the diffusion-weighted images than on the FLAIR images. This discrepancy was likely caused by a difference in lesion conspicuity between the sequences. Although a quantitative evaluation was not performed, hyperintense intraventricular debris and pus were more conspicuous on the diffusion-weighted images than on the FLAIR images. In terms of visual inspection, diffusion-weighted imaging might provide better lesion contrast than FLAIR imaging for the detection of intraventricular debris and pus (Figs. 2A, 2B, 2C, 2D, 2E, 3A, 3B, 3C, 3D, 3E, 3F, and 3G).

View larger version (131K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 2A —2-month-old boy with acute pyogenic meningitis. T2-weighted image shows slight dilatation of left occipital horn of lateral ventricle without findings of intraventricular and periventricular lesion.

View larger version (137K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 2B —2-month-old boy with acute pyogenic meningitis. FLAIR image shows small area of slight hyperintensity in left occipital horn and slight hyperintensity along ventricular wall of left occipital horn. Leptomeningeal hyperintensity is also noted in left frontal region, suggesting meningitis.

View larger version (115K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 2C —2-month-old boy with acute pyogenic meningitis. Contrast-enhanced T1-weighted image shows no significant enhancement of ventricular lining compared with meningeal enhancement in left frontal region, which is consistent with meningitis.

View larger version (116K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 2D —2-month-old boy with acute pyogenic meningitis. Diffusion-weighted image reveals areas of intraventricular hyperintensity in bilateral occipital horn. Right intraventricular lesion is not detected on other sequence images. Periventricular abnormal signal intensity is indeterminate.

View larger version (98K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 2E —2-month-old boy with acute pyogenic meningitis. Apparent diffusion coefficient (ADC) map shows decreased ADC values in areas corresponding to lesions on diffusion-weighted image (D).

View larger version (148K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 3A 39-year-old man with AIDS and intracranial tuberculous infection. FLAIR image shows intraventricular lesions in left occipital horn of lateral ventricles with areas of hyperintensity along ventricular lining. Alteration of signal intensity in right occipital horn is minimal. Areas of hyperintensity suggesting infarction of bilateral basal ganglia and right frontal region are also seen.

View larger version (132K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 3B 39-year-old man with AIDS and intracranial tuberculous infection. Contrast-enhanced T1-weighted image shows areas of slight hyperintensity relative to CSF in bilateral occipital horn. No periventricular enhancement is seen.

View larger version (120K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 3C 39-year-old man with AIDS and intracranial tuberculous infection. Diffusion-weighted image reveals bilateral hyperintense intraventricular lesions with left predominance.

View larger version (135K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 3D 39-year-old man with AIDS and intracranial tuberculous infection. FLAIR image 10 days after initial study shows newly developed cerebral lesion in left occipital region adjacent to left occipital horn in which intraventricular lesions have persisted. Areas of hyperintensity along ventricular wall are prominent. Dilated lateral ventricles and third ventricle are also newly seen.

View larger version (139K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 3E 39-year-old man with AIDS and intracranial tuberculous infection. Contrast-enhanced T1-weighted image 10 days after initial study shows focal enhancement in left occipital region, suggesting cerebritis. No enhancement of ventricular wall is noted.

View larger version (129K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 3F 39-year-old man with AIDS and intracranial tuberculous infection. Diffusion-weighted image 10 days after initial study shows areas of conspicuous focal hyperintensity in left occipital horn and along ventricular wall of bilateral occipital horn, which are different from other sequences in distribution. Hyperintensity in left occipital region suggests newly developed cerebritis.

View larger version (99K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 3G 39-year-old man with AIDS and intracranial tuberculous infection. Diffusion-weighted image obtained at pons level 10 days after initial study shows hyperintensity in dependent portion of dilated fourth ventricle and in right cerebellopontine cistern, suggesting intraventricular debris and pus and meningitis, respectively.

FLAIR imaging had the highest detectability rate for periventricular abnormalities, although this MRI finding is less frequent than intraventricular debris and pus. When symmetric abnormalities are present, the possibility of a periventricular inflammatory process should be considered. Areas of hyperintensity are frequently observed on FLAIR images of the periventricular white matter, particularly around the anterior and posterior horns of the ventricles. These areas of hyperintensity are called "caps and rims" and are considered to suggest age-related chronic ischemic changes in the white matter that are histologically characterized by myelin pallor, gliosis, and arteriosclerosis in the caps and subependymal gliosis and loss of the ependymal lining in the rims [6].

Furthermore, in the patients with hydrocephalus, the FLAIR images occasionally showed areas of periventricular hyperintensity in a circumferential fashion caused by the transependymal migration of CSF [3]. Interestingly, contrast-enhanced T1-weighted imaging was less sensitive at detecting ventricular wall abnormalities compared with the FLAIR and T2-weighted images. One possible explanation for this discrepancy is that false-positive abnormal periventricular findings on the FLAIR and T2-weighted images may have been included among the positive cases or that ventricular wall enhancement arising from the breakdown of the blood-brain barrier in the ependymal region may occur during a relatively advanced phase of ventriculitis.

Possible routes through which a pathogen might enter the intraventricular system include direct implantation secondary to trauma and postsurgical conditions, such as ventricular catheter placement; contiguous extension, such as the rupture of a brain abscess and extension into the ventricles; and hematogenous spread to the subependyma or the choroid plexus [2, 4, 7]. We postulated that the backflow of CSF from the extraventricular spaces into the intraventricular space might be another possible route of infection leading to ventriculitis. In three MRI studies in three patients, we observed abnormal intensities located in the fourth ventricle that had signal features identical to those of the intraventricular debris and pus seen in the lateral ventricles in the present study (Fig. 3G). We speculated that the debris and pus may have been conveyed from the lateral ventricles craniocaudally or from the extraventricular spaces caudocranially along with the CSF flow. In one of these patients, an additional follow-up MRI study revealed the development of cerebritis in the posterior lobe adjacent to the posterior horn of the lateral ventricle, in which the intraventricular debris and pus had persisted since the time of the initial study (Figs. 3A, 3B, 3C, 3D, 3E, 3F, and 3G). This finding suggests that the implantation of intraventricular debris and pus in the ventricular wall might cause ependymitis, leading to cerebritis or the formation of a brain abscess.

A blood clot in the ventricular system, caused by CSF backflow, is sometimes observed in cases of subarachnoid hemorrhage. This phenomenon is thought to arise from the alteration of pulsatile CSF flow dynamics by reduced compliance of the intracranial subarachnoid spaces as a result of the hemorrhage [8]. Other abnormalities of the subarachnoid space, including infection, might also contribute to changes in CSF flow. This concept might explain the observation that ventriculitis is often associated with meningitis.

Choroid plexitis is thought to be another MRI finding associated with ventriculitis. Imaging findings for choroid plexitis have been well described and include a poorly defined margin of a swollen choroid plexus and contrast enhancement of the choroid plexus [9]. We observed choroid plexitis on five contrast-enhanced T1-weighted scans obtained in four patients. In each case, the MR images showed asymmetric enlargement of the choroid plexus. However, abnormal choroid plexus findings could not be identified with certainty on the diffusion-weighted and FLAIR images, possibly because the abnormal intensities produced by the intraventricular debris and pus were too bright, making it difficult to differentiate them from findings indicating choroid plexitis.

The present study has at least three major limitations. First, the number of patients was small. However, this small series of patients diagnosed as having ventriculitis over a 5-year period reflects the rarity of this disease. Second, the presence of a causative organism in CSF drained directly from the lateral ventricles was not confirmed, except in three patients who underwent surgical interventions because of their postsurgical status for other indications (tumor resection, aneurysm clipping, and existing ventriculostomy tube for shunt). Although prompt treatment of ventriculitis is critical, surgical intervention has not been established as a treatment of first choice, especially for early stage ventriculitis [10]. We think that the laboratory findings for CSF obtained from a lumbar puncture, the clinical follow-up information, and the previously reported MRI features were sufficient to diagnose ventriculitis in the present study. Third, our study had a selection bias because cases with false-negative MRI findings were not included. However, if patients with ventriculitis have negative MRI findings, it is difficult to diagnose ventriculitis based only on clinical findings. No false-positive cases occurred in the present study, possibly implying that MRI contributes significantly to the diagnosis of ventriculitis.

In conclusion, diffusion-weighted and FLAIR imaging may be valuable MRI sequences for detecting intraventricular debris and pus, suggestive of ventriculitis. Diffusion-weighted imaging may be particularly useful for recognizing intraventricular debris and pus because of the conspicuity of the lesions, drawing attention to the existence of ventriculitis. FLAIR imaging might be more valuable than contrast-enhanced T1-weighted imaging for depicting periventricular abnormalities, the second most common MRI feature of ventriculitis.

References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Pezzullo JA, Tung GA, Mudigonda S, Rogg JM. Diffusion-weighted MR imaging of pyogenic ventriculitis. AJR2003; 180:71 -75[Abstract/Free Full Text]
2. Kanamalla US, Ibarra RA, Jinkins JR. Imaging of cranial meningitis and ventriculitis. Neuroimaging Clin N Am2000; 10:309 -331[Medline]
3. Fukui MB, Williams RL, Mudigonda S. CT and MR imaging features of pyogenic ventriculitis. Am J Neuroradiol2001; 22:1510 -1516[Abstract/Free Full Text]
4. Barloon TJ, Yuh WT, Knepper LE, Biller J, Ryals TJ, Sato Y. Cerebral ventriculitis: MR findings. J Comput Assist Tomogr 1990;14:272 -275[Medline]
5. Rana S, Albayram S, Lin DD, Yousem DM. Diffusion-weighted imaging and apparent diffusion coefficient maps in a case of intracerebral abscess with ventricular extension. Am J Neuroradiol2002; 23:109 -112[Abstract/Free Full Text]
6. Chimowitz MI, Estes ML, Furlan AJ, Awad IA. Further observations on the pathology of subcortical lesions identified on magnetic resonance imaging. Arch Neurol1992; 49:747 -752[Abstract]
7. Lambo A, Nchimi A, Khamis J, Lenelle J, Francotte N. Primary intraventricular brain abscess. Neuroradiology2003; 45:908 -910[CrossRef][Medline]
8. de Marco G, Idy-Peretti I, Didon-Poncelet A, Baledent O, Onen F, Feugeas MC. Intracranial fluid dynamics in normal and hydrocephalic states: systems analysis with phase-contrast magnetic resonance imaging. J Comput Assist Tomogr 2004;28:247 -254[CrossRef][Medline]
9. Mathews VP, Smith RR. Choroid plexus infections: neuroimaging appearances of four cases. Am J Neuroradiol1992; 13:374 -378[Medline]
10. Takeshita M, Kawamata T, Izawa M, Hori T. Prodromal signs and clinical factors influencing outcome in patients with intraventricular rupture of purulent brain abscess. Neurosurgery2001; 48:310 -316[CrossRef][Medline]

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This Article Abstract Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Fujikawa, A. Articles by Nitatori, T. Search for Related Content PubMed PubMed Citation Articles by Fujikawa, A. Articles by Nitatori, T. Social Bookmarking                      What's this?