QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted 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 HighWire Citing Articles via Google Scholar Google Scholar Articles by Duerstock, B. S. Articles by Borgens, R. B. Search for Related Content PubMed PubMed Citation Articles by Duerstock, B. S. Articles by Borgens, R. B.

Three-dimensional morphometry of spinal cord injury following polyethylene glycol treatment

Bradley S. Duerstock and Richard B. Borgens^*

Center for Paralysis Research, Institute for Applied Neurology, Department of Basic Medical Sciences, School of Veterinary Medicine, Purdue University, West Lafayette, IN 47907-1244, USA

View larger version (100K): [in a new window]   Fig. 1. Three-dimensional reconstruction of a normal spinal cord segment. An uninjured spinal cord segment was reconstructed using the isocontouring method to demonstrate that a familiar, normal morphology is reproduced. This serves as an image ‘control’ or reference for the reader since spinal cord injury produces an unfamiliar, unusually stenotic, and misshapen structure. The white matter was made transparent while the central gray matter was rendered in green in this visualization. The dorsal and ventral surfaces are flat because these very small sections (cutting into and out of the tissue block) were usually lost or damaged. In this and subsequent figures, the cylindrical icon indicates the relative orientation of the three-dimensional segment. The blue color indicates the caudal end of the cylinder. Unless noted otherwise, the dorsal surface of all reconstructed spinal cords faces towards the top of the page.

View larger version (132K): [in a new window]   Fig. 2. Histological sections of the lesion in PEG-treated and untreated spinal cords. Typical horizontal/longitudinal histological sections acquired to computer for three-dimensional reconstruction. All views shown were obtained from histological sections at the epicenter or most extensively damaged region of the lesion. A and C are typical of untreated (control) spinal cord lesions, while B and D are representative of those treated with PEG. Note the extensive cystic cavitation in control cords and the relative absence of cavitation in the PEG-treated spinal cords. In A and C, the rostral/caudal extents of the cysts are beyond the boundaries of the image. N, intact spinal cord parenchyma; L, lesion; Cy, cysts. The rostral end is to the left in all images. Scale bar (for A–D), 1 mm.

View larger version (65K): [in a new window]   Fig. 3. Quantitative querying of three-dimensional reconstructions. By selecting different isovalues, specific structures of interest can be imaged three-dimensionally (refer to Materials and methods). In A, the 4.2 mm length of reconstructed control spinal cord is shown containing the entire region of damaged spinal cord, including cysts and lesioned parenchyma. The three-dimensional visualization algorithm also allowed the surface area and volume of this structure to be determined. The shape of the segment is compacted because of compression injury, although the flattened dorsal and ventral surfaces are artifacts arising from the loss of a few histological sections during sectioning as explained in Fig. 1 and in Materials and methods. In B, the intact and damaged parenchyma of the spinal cord segment are shown without the cystic cavitations, which could be imaged separately (by isovalue selection) and deleted from the overall reconstruction. Note the presence of a large cyst extending from the center of the rostral end of the segment (facing the viewer) as a pocket within the spinal cord. The surface area and volume of the cysts were easily calculated by subtracting numerical values obtained from this reconstruction (B) from those obtained from the section shown in A. (C) The intact spinal cord parenchyma, which was also evaluated quantitatively. Note that most of the spared parenchyma is found at the periphery of the spinal segment. Numerical values derived from this reconstruction (C) were subtracted from those for B to give the surface area and volume of the lesion. The cylindrical icon gives the relative orientation of the three-dimensional segment; gray is the rostral end and black is the caudal end. Scale bar (for A–C), 1 mm.

View larger version (88K): [in a new window]   Fig. 4. Three-dimensional reconstruction of a control spinal cord segment. (Ai) A three-dimensional reconstruction of an entire 4.2 mm long spinal cord segment fenestrated by cysts. This image displays both normal and damaged spinal cord tissue with the cysts removed. One can see through this image because cystic cavitations transverse the width of the cord. Cysts can also be seen at the rostral end of the segment as pockets or invaginations at the end of the reconstructed spinal segment where these spaces (filled with fluid in situ) have extended beyond the boundary of the segment sectioned (refer to Fig. 2A). In Bi and Ci, only the undamaged parenchyma has been reconstructed. Aii shows cavitated and hemorrhagic gray matter inundated with macrophages (large circular cells in the center of the image). Bii shows a low-power photomicrograph of a region of partial sparing at the boundary of gray and white matter. Note the three well-stained neurons. Subpial sparing of white matter axons is shown in Cii. Note the orientation in Bi. In Ci, the spinal segment was rotated by approximately 200° in the horizontal plane. These reconstructions emphasize that intact parenchyma was mostly restricted to the periphery of the injured spinal segment and show the absence of contiguous gray and white matter through the lesion. The cylindrical icon gives the relative orientation of the three-dimensional segment; gray is the rostral end and black is the caudal end. Scale bars, 1 mm.

View larger version (61K): [in a new window]   Fig. 5. Three-dimensional reconstructions of immediate (A) and delayed (B) PEG-treated spinal cord segments. In A, the histological data set was obtained from an animal to which PEG had been applied within 30 min of injury. (B) Three-dimensional reconstructions of a spinal cord segment for which application of PEG to the injury site was delayed for approximately 7 h. As in Fig. 4, the top images show both the normal and damaged parenchyma with the cysts removed from the image. Both spinal cord segments were markedly compressed at the epicenter of the compression injury. In the middle and bottom images, the intact parenchyma of the spinal segments is imaged. Bottom, the spinal segments have been rotated by approximately 180° in the horizontal plane from the middle image. These views emphasize the greater extent of contiguous gray and white matter through the lesions of PEG-treated cords compared with controls (see also Fig. 4). The cylindrical icon gives the relative orientation of the three-dimensional segment; gray is the rostral end and black is the caudal end. Scale bar, 1 mm

View larger version (66K): [in a new window]   Fig. 6. Three-dimensional reconstruction of cavitation in control (A) and PEG-treated (B) spinal cord segments. In A, the dorsal surface of the spinal cord segment is facing the viewer; the ventral surface is towards the page. The surrounding tissues were made transparent, and the most dorsal surface of this spinal cord segment was also removed to provide an unobstructed view of the cysts (shown in dark red). (A) Top image, a large cystic cavity occupies the rostral (left) half of this segment. In the middle image, the spinal segment has been rotated vertically by approximately 90° so that the dorsal aspect is towards the top of the page, showing a side view of this extensive cavitation. Bottom image, a second sham-treated spinal cord segment shown for comparison. Note the similar, and very large, cyst in the center of the segment (colored brown). The ventral surface is facing the top of the page. (B) Top image, cysts were dispersed throughout the PEG-treated segment but mainly localized on either side of the site of injury. Middle image, this spinal segment has been rotated vertically so that the dorsal aspect is towards the top of the page, showing a side view of cavitation. Bottom image, an oblique side view of a second PEG-treated spinal cord segment; the dorsal surface is towards the top of the page. The embedded cysts are imaged in a darker color than the parenchyma surrounding them. These spinal cord reconstructions emphasize the markedly decreased amount of cavitation in PEG-treated cords compared with controls. In the orientation icon, red is the rostral end and blue is the caudal end. Scale bar, 1 mm.

View larger version (24K): [in a new window]   Fig. 7. Comparisons of volume and surface area measurements between untreated and PEG-treated spinal cords. (A) The normalized volume measurements for intact parenchyma, lesion and cystic cavities. The volume measurements for intact parenchyma and cavitation were statistically significantly different (asterisks) between the control group and both PEG-treated groups. (B) The normalized surface area measurements for the same groups. The surface area measurements for lesions were statistically significantly different between the control group and both PEG-treated groups. There was a statistically significant difference in the amount of cavitation between the immediate and delayed PEG-treated groups. Values are means + S.D. *Indicates a significant difference between control and experimental spinal cords (P<0.05; unpaired Student’s t-test). Indicates a significant difference between immediate and delayed PEG-treated spinal cords (P<0.05; unpaired Student’s t-test).

View larger version (30K): [in a new window]   Fig. 8. Correlation tests between the volume of intact spinal cord tissue, cavitation and the lesion. Pearson correlation tests were performed to determine whether there were linear relationships between the volume percentages of intact parenchyma, lesion and cavitation. (A) There appears to be a strong inverse relationship between the volume of intact nervous tissue and the volume of lesion in the untreated control group and in the immediate and delayed PEG-treated groups. Thus, an increase in the volume of spinal cord sparing translates into a decrease in lesion formation. (B) A correlation between the volumes of undamaged tissue and the volume of cysts was observed only in the delayed PEG-treated group. (C) The amount of cavitation appears to be related to the amount of lesion being formed in the experimental spinal cords. Note that the correlation test for only the delayed PEG-treated group was statistically significant (P<0.05).