The Cavum Septi Pellucidi in Term and Preterm Newborn Infants Stephen
H. Mott, MD;
John
B.
Bodensteiner, MD; Walter C. Allan, MD
Abstract We reviewed cranial sonographic studies done on 108 normal newborn infants to determine the prevalence and variability in size of the cavum septi pellucidi (CSP). Infants were classified according to gestational age by 2-week intervals. At 24 weeks, only four normal scans were identified. Between 26 and 34 weeks, ten consecutive normal scans were used since all infants had a CSP. At 36, 38, and 40 weeks, all normal scans were counted in order to obtain a prevalence estimate. A CSP was seen in all normal infants below 36 weeks of gestational age. At 36, 38, and 40 weeks, only 69%, 54%, and 36%, respectively, had CSPs. There was no significant change in the width, depth, or area of the CSP with age, birth weight, or biparietal diameter. A CSP greater than 0.95 cm in width or greater than 0.81 cm in depth is outside the normal range and may represent anomalous development of the midline structures of the brain. ( J Child Neurol 1992;7:35-38).
consists of two leaves of tissue which extend from the corpus callosum to the columns of the fornix and separate the lateral ventricles. The outgrowth of the banks of the median groove of the massa commissuralis become the lamellae of the septum.l2 These lamellae are separate in utero and fuse as the fetal brain matures. In the fetus, the closed (cavum) septum pellucidum (CSP) forms when the rostrum of the corpus callosum becomes complete. Thus, the formation of the septum pellucidum is intimately related to the formation of the midline cerebral structures, particularly the corpus callosum. The frequency with which a CSP persists into extrauterine life has been studied by various methods including autopsy, pneumoencephalography, and computed tomography (CT) (Table 1). In one autopsy study, a CSP was found in all fetuses, in 97% of infants less than 1 week old, and in 15% of infants at 6 months of age.4 A retrospective study of
The septum pellucidum
Received May 24, 1991. Received revised August 2, 1991. Acfor publication August 5, 1991. From the Department of Pediatrics (Dr Mott) and Neurology (Dr Allan), Maine Medical Center, Portland, ME, and the Departments of Neurology and Pediatrics (Dr Bodensteiner), West Virginia University Health Science Center, Morgantown, WV. Presented in part at the 19th annual meeting of the Child Neurology Society, Atlanta, GA, October 19, 1990. Address correspondence to Dr Bodensteiner, Department of Neurology, West Virginia University Health Science Center, Room G-103, Morgantown, WV 26506.
cepted
CT material found a CSP in 3.9% of infants less than 1 year of age and 2.2% of children over 10 years of age.g These studies suggest that the frequency of CSP varies depending on the age and nature of the population and the method of study employed. We studied the prevalence and size of the CSP in normal, living, term and preterm newborn infants using cranial sonographic imaging.
Patients and Methods All newborn infants less than 35 weeks’ gestational age admitted to the Maine Medical Center Neonatal Intensive Care Center are studied routinely by cranial sonographic examination. Infants 35 weeks’ gestational age or older were scanned when clinically indicated. We studied infants with cranial sonographic examinations performed between January 1, 1985, and December 31, 1988. All scans were done within the first week of life. Various scanners were used over the years sampled, but all were mechanical sector scanners and either 5-mHz or 7.5-mHz transducers were used. All scans were performed in the coronal and sagittal planes and archived on videotape. The sonographic scan reports were reviewed and normal scans selected for the present study. The videotape of the scans selected were reviewed, and the measurements of the cavum septum pellucidum and biparietal diameter of the skull were made. The birth weight was determined from the medical record. The width and depth of the CSPs were measured using the internal scale markers on the real-time sonographic image as a calibration standard. The scan most appropriate for these measurements is in the near perpendicular coronal plane just anterior to the level of the foramen of Monro. The area of the CSP was calculated
35
Downloaded from jcn.sagepub.com at RMIT UNIVERSITY on March 15, 2015
TABLE 1 Previous Studies of the Cavum
CSP
=
raphy ;
septi pellucidi; MGA computed tomography.
cavum
CT
=
Septi Pellucidi
=
months
gestational
age; PEG
=
pneumoencephalog-
from the width and depth measurements made in the coronal plane (Figure 1). Only four normal scans could be identified in infants at 24 weeks of gestational age. Between 26 and 34 weeks’ gestational age, ten consecutive normal scans were used, since all infants had CSP at these gestational ages. At gestational ages 36, 38, and 40 weeks, all normal scans were used, in order to obtain a prevalence estimate, as well as an accurate estimate of average size of the CSP. The width, depth and area of the CSP in relationship to the gestational age of the infant were compared using regression analysis. The width, depth, and area of the CSP were also compared to the biparietal diameter of the head and birth weight using regression analysis.
Results All infants less than 36 weeks of gestational age had a CSP. The prevalence of CSP decreased as the infants approached term, being present in only 36% of normal full-term newborn infants (Table 2). Based on regression analysis, there was no significant correlation between gestational age and the width, depth, or area of the CSP. There was no significant correlation between the biparietal diameter of the head and the size of the CSP based on regression analysis. Finally, there was also no significant correlation between the width, depth, and area of the CSP and the birth weight of the infants. Since there was no correlation between gestational age and the size of the CSP, the data for all ages could be pooled for purposes of determining the acceptable limits for the size of the CSP (Figure 2). The mean width of the CSP was 0.57 cm, with a 0.95 standard deviation of 0.19 cm (mean + 2 SD with was 0.53 the The mean of CSP cm, cm). depth a standard deviation of 0.14 cm (mean + 2 SD 0.81 cm). =
=
FIGURE 1
Coronal plane sonogram of typical infant of 36 weeks’ gestational age, demonstrating the cavum septi pellucidi.
Discussion When identified in children or adults, the CSP has generally been considered a normal variant because of the apparent frequency of occurrence and the fact that no distinctive clinical syndrome has been identified in patients with this finding. The true prevalence of this finding in the general population remains undetermined. The studies available clearly represent populations with variable degrees of bias introduced by the method of ascertainment. Although the criteria for and the frequency of autopsy may have been more liberal in the past, the prevalence of CSP in the population of individuals who
36
Downloaded from jcn.sagepub.com at RMIT UNIVERSITY on March 15, 2015
TABLE 2 Results
*Mean ± SD. CSP = cavum
septi pellucidi.
have been autopsied still may not reflect the prevalence in the general living population. Similarly, the indications for pneumoencephalography were such that one would clearly not equate the population of individuals undergoing pneumoencephalography with a normal population. The selection bias inherent in the previous prevalence studies would account for the difference in frequency estimates among the individual studies and between our study and the previous studies. Our study confirms the finding that premature infants all have a CSP.3~4 Furthermore, the prevalence of CSP decreases as the fetus approaches term, so that at 38 weeks of gestational age 56% will have a cavum septum pellucidum, and at 40 weeks of gestational age only 36% of normal infants will demonstrate a persistent CSP.
Although the size of the CSP did not correlate gestational age, biparietal diameter, or birth weight, the variation in size observed (measured by the standard deviation of the mean) was relatively small. Accordingly, it is possible to conclude that an with the
infant who demonstrates a CSP greater than two standard deviations above the mean width (0.95 cm) or mean depth (0.81 cm) is outside the range of normal variability. A cavum of such dimensions may represent an aberration of maturation or development of the midline structures of the brain and as such may have functional consequences.9 Although it would be possible to perform CT or magnetic resonance imaging scans on a normal population prospectively to determine the prevalence of CSP, to date no such studies have been reported. The prevalence of CSP in the normal postnatal population, therefore, remains undetermined. However, this prevalence figure is essential to the interpretation of any studies of this finding in various patient populations. 10,11 When a CSP is identified in the newborn period, it probably represents a variation of normal maturation, providing it is within the acceptable range of variation in size.
References
FIGURE 2 Graph of the mean (and standard deviation of the mean) width of the cavum septi pellucidi as a function of gesta-
tional age.
1. Rakic P, Yakovlev PI: Development of the corpus callosum and cavum septi in man. J Comp Neurol 1968;132:45-72. 2. Muller F, O’Rahilly R: Mediobasal prosencephalic defects, including holoprosencephaly and cyclopia, in relation to the development of the human forebrain. AmJ Anat 1989;185: 391-414. 3. Larroche JC, Baudley J: Cavum septi lucidi, cavum vergae, cavum veli interpositi: Cavities de la ligne mediane. Biol Neonatol 1961;3:193-236. 4. Shaw CM, Alvord EC: Cava septi pellucidi et vergae: Their normal and pathological states. Brain 1969;92:213-224.
37
Downloaded from jcn.sagepub.com at RMIT UNIVERSITY on March 15, 2015
5. Schunk H: 6.
Congenital dilatations of the septum pellucidum. Radiology 1963;81:610-618. Finke J, Koch G: Das Cavum septi pellucidi: Vorkommen und Aussagewert. Bericht über 128 Falle. Deutsch Z Nervenheilk
9. Bodensteiner JB, Schaefer GB: The wide cavum septum pellucidum : A marker of disturbed development. Pediatr Neurol
1990;6:391-394. 10.
1968;193:154-157. 7. Thieffry S, Lefèbvre J, Lepintre MJ, Fauré C, Masselin MS: Contribution à L’étude radiologique des malformations du plan sagittal interhémisphérique à propos de 45 observations. Acta Radiol
1958;50:242-252. 8. Nakano S, Hojo H, Kataoka K, Yamasaki S: Age related incidence of cavum septum pellucidum and cavum vergae on CT scans
of
pediatric patients.
Bodensteiner JB, Cowan L, Higgins WL: The septum pellucidum: An MRI study of prevalence and clinical associations in a pediatric population.J Neuroimaging,
Breeding LM, cavum
11.
1991;1:115-118. Bogdanoff B, Natter HM: Incidence of dum in adults: A
1989;39:991-992.
J Comput Assist Tomogr
1981;5:348-349.
38
Downloaded from jcn.sagepub.com at RMIT UNIVERSITY on March 15, 2015
sign
of boxer’s
cavum septum pelluciencephalopathy. Neurology
H. Mott, MD;
John
B.
Bodensteiner, MD; Walter C. Allan, MD
Abstract We reviewed cranial sonographic studies done on 108 normal newborn infants to determine the prevalence and variability in size of the cavum septi pellucidi (CSP). Infants were classified according to gestational age by 2-week intervals. At 24 weeks, only four normal scans were identified. Between 26 and 34 weeks, ten consecutive normal scans were used since all infants had a CSP. At 36, 38, and 40 weeks, all normal scans were counted in order to obtain a prevalence estimate. A CSP was seen in all normal infants below 36 weeks of gestational age. At 36, 38, and 40 weeks, only 69%, 54%, and 36%, respectively, had CSPs. There was no significant change in the width, depth, or area of the CSP with age, birth weight, or biparietal diameter. A CSP greater than 0.95 cm in width or greater than 0.81 cm in depth is outside the normal range and may represent anomalous development of the midline structures of the brain. ( J Child Neurol 1992;7:35-38).
consists of two leaves of tissue which extend from the corpus callosum to the columns of the fornix and separate the lateral ventricles. The outgrowth of the banks of the median groove of the massa commissuralis become the lamellae of the septum.l2 These lamellae are separate in utero and fuse as the fetal brain matures. In the fetus, the closed (cavum) septum pellucidum (CSP) forms when the rostrum of the corpus callosum becomes complete. Thus, the formation of the septum pellucidum is intimately related to the formation of the midline cerebral structures, particularly the corpus callosum. The frequency with which a CSP persists into extrauterine life has been studied by various methods including autopsy, pneumoencephalography, and computed tomography (CT) (Table 1). In one autopsy study, a CSP was found in all fetuses, in 97% of infants less than 1 week old, and in 15% of infants at 6 months of age.4 A retrospective study of
The septum pellucidum
Received May 24, 1991. Received revised August 2, 1991. Acfor publication August 5, 1991. From the Department of Pediatrics (Dr Mott) and Neurology (Dr Allan), Maine Medical Center, Portland, ME, and the Departments of Neurology and Pediatrics (Dr Bodensteiner), West Virginia University Health Science Center, Morgantown, WV. Presented in part at the 19th annual meeting of the Child Neurology Society, Atlanta, GA, October 19, 1990. Address correspondence to Dr Bodensteiner, Department of Neurology, West Virginia University Health Science Center, Room G-103, Morgantown, WV 26506.
cepted
CT material found a CSP in 3.9% of infants less than 1 year of age and 2.2% of children over 10 years of age.g These studies suggest that the frequency of CSP varies depending on the age and nature of the population and the method of study employed. We studied the prevalence and size of the CSP in normal, living, term and preterm newborn infants using cranial sonographic imaging.
Patients and Methods All newborn infants less than 35 weeks’ gestational age admitted to the Maine Medical Center Neonatal Intensive Care Center are studied routinely by cranial sonographic examination. Infants 35 weeks’ gestational age or older were scanned when clinically indicated. We studied infants with cranial sonographic examinations performed between January 1, 1985, and December 31, 1988. All scans were done within the first week of life. Various scanners were used over the years sampled, but all were mechanical sector scanners and either 5-mHz or 7.5-mHz transducers were used. All scans were performed in the coronal and sagittal planes and archived on videotape. The sonographic scan reports were reviewed and normal scans selected for the present study. The videotape of the scans selected were reviewed, and the measurements of the cavum septum pellucidum and biparietal diameter of the skull were made. The birth weight was determined from the medical record. The width and depth of the CSPs were measured using the internal scale markers on the real-time sonographic image as a calibration standard. The scan most appropriate for these measurements is in the near perpendicular coronal plane just anterior to the level of the foramen of Monro. The area of the CSP was calculated
35
Downloaded from jcn.sagepub.com at RMIT UNIVERSITY on March 15, 2015
TABLE 1 Previous Studies of the Cavum
CSP
=
raphy ;
septi pellucidi; MGA computed tomography.
cavum
CT
=
Septi Pellucidi
=
months
gestational
age; PEG
=
pneumoencephalog-
from the width and depth measurements made in the coronal plane (Figure 1). Only four normal scans could be identified in infants at 24 weeks of gestational age. Between 26 and 34 weeks’ gestational age, ten consecutive normal scans were used, since all infants had CSP at these gestational ages. At gestational ages 36, 38, and 40 weeks, all normal scans were used, in order to obtain a prevalence estimate, as well as an accurate estimate of average size of the CSP. The width, depth and area of the CSP in relationship to the gestational age of the infant were compared using regression analysis. The width, depth, and area of the CSP were also compared to the biparietal diameter of the head and birth weight using regression analysis.
Results All infants less than 36 weeks of gestational age had a CSP. The prevalence of CSP decreased as the infants approached term, being present in only 36% of normal full-term newborn infants (Table 2). Based on regression analysis, there was no significant correlation between gestational age and the width, depth, or area of the CSP. There was no significant correlation between the biparietal diameter of the head and the size of the CSP based on regression analysis. Finally, there was also no significant correlation between the width, depth, and area of the CSP and the birth weight of the infants. Since there was no correlation between gestational age and the size of the CSP, the data for all ages could be pooled for purposes of determining the acceptable limits for the size of the CSP (Figure 2). The mean width of the CSP was 0.57 cm, with a 0.95 standard deviation of 0.19 cm (mean + 2 SD with was 0.53 the The mean of CSP cm, cm). depth a standard deviation of 0.14 cm (mean + 2 SD 0.81 cm). =
=
FIGURE 1
Coronal plane sonogram of typical infant of 36 weeks’ gestational age, demonstrating the cavum septi pellucidi.
Discussion When identified in children or adults, the CSP has generally been considered a normal variant because of the apparent frequency of occurrence and the fact that no distinctive clinical syndrome has been identified in patients with this finding. The true prevalence of this finding in the general population remains undetermined. The studies available clearly represent populations with variable degrees of bias introduced by the method of ascertainment. Although the criteria for and the frequency of autopsy may have been more liberal in the past, the prevalence of CSP in the population of individuals who
36
Downloaded from jcn.sagepub.com at RMIT UNIVERSITY on March 15, 2015
TABLE 2 Results
*Mean ± SD. CSP = cavum
septi pellucidi.
have been autopsied still may not reflect the prevalence in the general living population. Similarly, the indications for pneumoencephalography were such that one would clearly not equate the population of individuals undergoing pneumoencephalography with a normal population. The selection bias inherent in the previous prevalence studies would account for the difference in frequency estimates among the individual studies and between our study and the previous studies. Our study confirms the finding that premature infants all have a CSP.3~4 Furthermore, the prevalence of CSP decreases as the fetus approaches term, so that at 38 weeks of gestational age 56% will have a cavum septum pellucidum, and at 40 weeks of gestational age only 36% of normal infants will demonstrate a persistent CSP.
Although the size of the CSP did not correlate gestational age, biparietal diameter, or birth weight, the variation in size observed (measured by the standard deviation of the mean) was relatively small. Accordingly, it is possible to conclude that an with the
infant who demonstrates a CSP greater than two standard deviations above the mean width (0.95 cm) or mean depth (0.81 cm) is outside the range of normal variability. A cavum of such dimensions may represent an aberration of maturation or development of the midline structures of the brain and as such may have functional consequences.9 Although it would be possible to perform CT or magnetic resonance imaging scans on a normal population prospectively to determine the prevalence of CSP, to date no such studies have been reported. The prevalence of CSP in the normal postnatal population, therefore, remains undetermined. However, this prevalence figure is essential to the interpretation of any studies of this finding in various patient populations. 10,11 When a CSP is identified in the newborn period, it probably represents a variation of normal maturation, providing it is within the acceptable range of variation in size.
References
FIGURE 2 Graph of the mean (and standard deviation of the mean) width of the cavum septi pellucidi as a function of gesta-
tional age.
1. Rakic P, Yakovlev PI: Development of the corpus callosum and cavum septi in man. J Comp Neurol 1968;132:45-72. 2. Muller F, O’Rahilly R: Mediobasal prosencephalic defects, including holoprosencephaly and cyclopia, in relation to the development of the human forebrain. AmJ Anat 1989;185: 391-414. 3. Larroche JC, Baudley J: Cavum septi lucidi, cavum vergae, cavum veli interpositi: Cavities de la ligne mediane. Biol Neonatol 1961;3:193-236. 4. Shaw CM, Alvord EC: Cava septi pellucidi et vergae: Their normal and pathological states. Brain 1969;92:213-224.
37
Downloaded from jcn.sagepub.com at RMIT UNIVERSITY on March 15, 2015
5. Schunk H: 6.
Congenital dilatations of the septum pellucidum. Radiology 1963;81:610-618. Finke J, Koch G: Das Cavum septi pellucidi: Vorkommen und Aussagewert. Bericht über 128 Falle. Deutsch Z Nervenheilk
9. Bodensteiner JB, Schaefer GB: The wide cavum septum pellucidum : A marker of disturbed development. Pediatr Neurol
1990;6:391-394. 10.
1968;193:154-157. 7. Thieffry S, Lefèbvre J, Lepintre MJ, Fauré C, Masselin MS: Contribution à L’étude radiologique des malformations du plan sagittal interhémisphérique à propos de 45 observations. Acta Radiol
1958;50:242-252. 8. Nakano S, Hojo H, Kataoka K, Yamasaki S: Age related incidence of cavum septum pellucidum and cavum vergae on CT scans
of
pediatric patients.
Bodensteiner JB, Cowan L, Higgins WL: The septum pellucidum: An MRI study of prevalence and clinical associations in a pediatric population.J Neuroimaging,
Breeding LM, cavum
11.
1991;1:115-118. Bogdanoff B, Natter HM: Incidence of dum in adults: A
1989;39:991-992.
J Comput Assist Tomogr
1981;5:348-349.
38
Downloaded from jcn.sagepub.com at RMIT UNIVERSITY on March 15, 2015
sign
of boxer’s
cavum septum pelluciencephalopathy. Neurology
