Article Text
Abstract
Background and objective: Subependymal pseudocysts and choroid plexus cysts are seen in newborns on cerebral ultrasound. Clinicians are unsure whether these findings are related to an underlying disease which affects long-term outcome and requires medical intervention. In an attempt to establish the diagnostic value of cystic lesions on cerebral ultrasound and guide clinical management we searched the medical literature and performed a meta-analysis.
Methods: We performed a systematic literature review and summarised the data on the value of subependymal pseudocysts or choroid plexus cysts for the diagnosis of chromosomal anomalies or congenital infections. Sensitivity, specificity, predictive values and likelihood ratios were calculated for single, multiple, unilateral and bilateral cysts.
Results: 305 patients with cystic lesions were retrieved. Bilateral cysts, irrespective of their number, had a sensitivity of 88% and negative predictive value of 94% for a congenital infection or genetic disorder. Unilateral single cysts had a specificity of 92% for normal microbiological and genetic results. Bilateral multiple subependymal pseudocysts or choroid plexus cysts had a positive likelihood ratio of 9.1 for a chromosomal anomaly or congenital infection. Unilateral cysts had a negative likelihood ratio of 0.2 for a congenital infection or chromosomal anomaly. There was a chance of 1 in 4–5 for a congenital infection or chromosomal anomaly if bilateral multiple subependymal pseudocysts or choroid plexus cysts were found.
Conclusions: Bilateral multiple subependymal pseudocysts or choroid plexus cysts suggest an underlying disease. Further investigations should be undertaken even if the patient is otherwise normal. Parents of well newborns with a single cyst should be reassured.
Statistics from Altmetric.com
With the improvements in ultrasound technology and the widespread use of cranial ultrasound assessment subependymal pseudocysts and choroid plexus cysts are increasingly seen in newborns on cerebral ultrasound scans. Postnatally choroid plexus cysts and subependymal pseudocysts have been identified in 1–5% of the newborn population on cranial ultrasound scans.1 2 The usefulness of screening newborns with subependymal pseudocysts or choroid plexus cysts for congenital infections or chromosomal anomalies is still under discussion. Some studies have suggested a relationship between laterality or number of cysts and the presence of an underlying pathology.3 4 However these findings are often incidental and the need for further investigations may be unclear. Although most cysts might exist in absence of associated anomalies, some will be markers of congenital infection or chromosomal anomalies with the likelihood of impaired neurodevelopmental outcome.2 5 6 7 8 Studies aiming to evaluate the diagnostic value of bilateral versus unilateral and single versus multiple cystic lesions as a guide to further investigations are not available. We performed a meta-analysis of the literature on choroid plexus cysts and subependymal pseudocysts to determine the diagnostic value of such cystic lesions in daily neonatal practice.
Methods
A structured database (Medline, EMBASE, Cinahl) literature search from 1966–2008 using the terms “subependymal cyst” OR “pseudocyst” OR “choroid plexus cyst” was performed. Thereafter, the references of all retrieved studies were checked for relevant citations. No language restrictions were used. Reports that did not include cerebral ultrasound as imaging technique were excluded from analysis. Included reports were analysed for site and number of cysts as well as confirmed associated congenital infections or chromosomal anomalies. Patients were grouped into different categories according to their ultrasound findings and presence of congenital infections or anomalies. Sensitivity, specificity, predictive values and likelihood ratios with their 95% confidence intervals (CIs) were calculated on the basis of the published data for the ultrasound findings and the corresponding microbiological/genetic results. Association between laterality or number of cystic lesions and presence of chromosomal anomaly or congenital infection was tested by calculation of φ coefficient of association and performance of the χ2 test of association. Significant correlation was assumed if p<0.001. Post-test probability was calculated based on an estimated combined prevalence of 3% for congenital infections and chromosomal anomalies according to data from the Health Protection Agency and Eurocat.9 10
Results
In total 22 original reports were included. All studies were published in English except for one which was published in German. Only three studies were published after the year 2000. There were 10 prospective case series/cohort studies, 11 retrospective case series/cohort studies and one meta-analysis. Fourteen included information on neurodevelopmental outcome with a variable degree of standardisation of the assessment tools. Four included postmortem studies. A total of 305 patients with cystic lesions were reported. The data of the selected studies are displayed in the following tables separated into either unilateral single or multiple cysts (table 1) and bilateral single or multiple cysts (table 2). Sensitivity, specificity, predictive values, likelihood ratios and post-test probability are presented in table 3.
There was a statistically significant correlation between the presence of bilateral multiple subependymal pseudocysts or choroid plexus cysts and the existence of chromosomal anomalies or congenital infections (φ 0.62, p<0.001). There was a marked difference in the sensitivity, specificity, predictive values and likelihood ratios for different diagnostic constellations: bilateral versus unilateral single or multiple cysts, multiple versus single unilateral or bilateral cysts and bilateral multiple cysts versus unilateral single cysts. The presence of bilateral cysts irrespective of their number had the highest sensitivity and negative predictive value for an underlying congenital infection or genetic disorder (88% and 94% respectively) (table 3). A unilateral single cyst had the highest specificity for normal microbiological and genetic results (92%) (table 3). Combining laterality and number of cysts revealed the highest positive likelihood ratio for a chromosomal anomaly or congenital infection if bilateral multiple subependymal pseudocysts or choroid plexus cysts were present (positive likelihood ratio 9.1) (table 3). Unilateral cysts had the lowest negative likelihood ratio for a congenital infection or chromosomal anomaly (negative likelihood ratio 0.2). The post-test probability of a congenital infection or chromosomal anomaly was 22% with bilateral multiple subependymal pseudocysts or choroid plexus cysts and 8% with either bilateral cysts irrespective of number or multiple cysts irrespective of laterality (table 3).
Discussion and conclusions
Mammalian choroid plexus develops at four sites in the roof of the neural tube shortly after its closure, in the order fourth, lateral and third ventricles.29 The choroid plexus epithelial cells represent a continuation of, and have the same origin as, ventricular ependymal cells, and are regarded as modified ependymal cells.30 Cystic lesions of the choroid plexus are thought to be epithelial indentations lining the interlobular clefts that may have become buried in the stroma during early development.31
Ventricular ependymal cells show marked proliferation during the embryonic period.32 During this time the germinal matrix zone is much larger than at term.33 Subependymal pseudocyst formation is the result of cellular destruction producing a residual cavity under the ependyma or far from it, depending on the original place of the lesion and the age of gestation.33 This cavity is lined with immature undifferentiated cells, sometimes bordered by thick glial meshwork.33 These characteristics do not fulfil the criteria of cysts and therefore the term pseudocyst is used for subependymal lesions.33
Although the histological mechanisms of cyst formation might be different, we did not distinguish between subependymal pseudocysts and choroid plexus cysts in this study. This is because both subependymal pseudocysts and choroid plexus cysts have been related to systemic disorders such as congenital infections or chromosomal anomalies.31 34 35 36
Stadlan and Gilles in 1967, as well as Larroche in 1972, described subependymal pseudocysts in postmortem examinations of newborns.25 28 33 The latter found cystic lesions unilaterally and bilaterally in different locations and in varying number.33 Seventeen of his 22 investigated cases were abnormal pregnancies with seven (32%) being related to congenital infections.33 Shaw reported in 1974 a rate of 23% of proven viral infections related to subpendymal pseudocysts in his postmortem studies.4 He highlighted the fact that bilateral cystic lesions seemed to be related to congenital infections.4
Since then these findings have been repeated in several reports pointing out that in particular bilateral intracranial subependymal pseudocysts might be due to congenital infections with neurotropic virus like cytomegalovirus or rubella virus, but also due to neonatal ventriculitis and lactic acid encephalopathy.5
Bilateral intracranial multilocular cysts have also been thought to be related to chromosomal anomalies: trisomy 18 and 21 are known to present with such anomalies as well as Cri du chat (5p-), Hirschorn-Wolf, Schinzel-Giedion and Zellweger Syndrome.2 5 37 38 39 40 In addition we saw a patient with tetrasomy 18 who presented with bilateral multiple subependymal pseudocysts in our population of patients. It is possible that other syndromes may have similar findings as well.
Intraventricular cysts have also been described after intraventricular hemorrhages (IVH) and it might be that the preceeding patho-mechanism is a germinal matrix bleed or as Schwartz termed it in 1961 “sequelae of subependymal vena terminalis haematomas”.33 41 His findings are in contrast to Shaw’s descriptions who could not confirm any evidence of previous haemorrhage in the cystic lesions.4
Despite all the described pathological causes a large proportion of subependymal pseudocysts and choroid plexus cysts may be an isolated finding without any clear underlying cause.19
The clinician therefore faces the question about the need for further investigations with the potential anxiety that this might mean for parents. So far publications regarding the clinical association of subependymal pseudocysts or choroid plexus cysts and chromosomal anomalies or congenital infections have not addressed the diagnostic value and reliability of such findings for daily clinical practice. We analysed the literature trying to answer the question whether such findings should immediately prompt confirmatory chromosomal analysis or microbiological screen. Our post-test probability results suggest that there is a chance of approximately 1 in 4–5 for an underlying chromosomal anomaly or congenital infection if bilateral multiple subependymal pseudocysts or choroid plexus cysts are found on cranial ultrasound. Appropriate investigations should be considered, especially if other risk factors, additional symptoms or signs of a congenital infection or chromosomal anomaly are present. Most of these newborns will show some mild to moderate abnormalities on neurological examination irrespective of other risk factors. Their significance for long-term neurological outcome is unclear.2 5 6 7 8 Hence if no investigations are undertaken follow-up is advisable and parents should be informed about the potential adverse long-term neurological outcome.
This meta-analysis has shown that the likelihood of a congenital infection or a chromosomal anomaly in association with a unilateral single cyst is negligible. Further investigations should not be necessary. Newborns with a single cyst represent the majority of patients according to our experience. It is therefore possible to avoid performing unnecessary blood and urinary sampling in the majority of infants with pseudocysts or choroid plexus cysts, especially in the absence of other obvious infection and genetics related history or clinical signs.
The position of the cysts has been related to the underlying aetiology and pathogenesis.42 It has been postulated in the literature that high-resolution sonography and the position of the cystic lesion would help distinguish between haemorrhagic and non-haemorrhagic germinolysis.42 43 Furthermore sonographic criteria have been published to distinguish between subependymal pseudocysts and choroid plexus cysts.44 Based on the above mentioned facts and our own experience we feel that this is not necessary. The increased skills and time required to detect the differences should not translate into different investigations or management plans as the sonographic findings result more likely from lesions of the same origin.
As patients were pooled from various studies in different countries with different demographics it was not possible to perform an accurate post-test probability calculation. However it was reassuring to see that our value of 22% for bilateral multiple cystic lesions and congenital infections or chromosomal anomalies was very similar to the figures quoted by Larroche (1972) and Shaw (1974), 34% and 23% respectively.4 33
In conclusion improvement in ultrasound technology and skills will certainly help pick up more of these sometimes difficult findings on scans raising the question about the clinical significance and necessity of additional investigations. Overall, it seems that the appearance of bilateral multiple subependymal pseudocysts or choroid plexus cysts should give enough reason for concern to undertake investigations even if the findings are incidental in an otherwise normal patient. If no diagnosis of underlying disease can be made neurodevelopmental follow-up is advisable as outcome may be unfavourable. Parents of well newborns with a single cyst should be reassured and neurodevelopmental follow-up is not necessary.
What is already known about this topic?
Choroid plexus cysts and subependymal pseudocysts are seen in 1–5% of the newborn population on cranial ultrasound scans.
There is controversy about the need for further microbiological and/or genetic investigations in newborns with such findings.
What this study adds?
One in 4–5 infants with bilateral multiple subependymal pseudocysts or choroid plexus cysts will have a congenital infection or genetic anomaly.
Parents of well newborns with a single cyst should be reassured, no investigations are necessary.
Acknowledgments
S Clayton, Clinical Librarian, Brighton & Sussex University Hospitals NHS Trust.
REFERENCES
Footnotes
Competing interests None.
Provenance and Peer review Not commissioned; externally peer reviewed.