J Neurosurg 51:352-354, 1979
Cerebrospinal fluid cytology after subarachnoid hemorrhage UMEO ITO, M.D., AND YUTAKA INABA,M.D.
Department of Neurosurgery, Tokyo Medical and Dental University, Tokyo, Japan t,' A method is described which has been found capable of detecting subarachnoid hemorrhage (SAH) up to 15 to 17 weeks after its occurrence. The episode of SAH was confirmed by bloody and/or xanthochromic cerebrospinal fluid (CSF) at the time of SAH onset. In this study, 47 samples of lumbar CSF from diagnostically confirmed SAH patients were used. The CSF cells were collected onto slides and stained with May-Gruenwald-Giemsa or Perl's reagent. Iron-positive cells were detected at I week, increased by 4 to 6 weeks to 8.5% of total nucleated cells, and decreased to 1% by 15 to 17 weeks. All 27 samples obtained at 2 to 9 weeks after SAH showed iron-positive cells. No iron-positive cells (false-negative samples) were noted in 25% (one of four) of samples obtained during the first week, and in 33% (one of three) of samples obtained 10 to 12 weeks and 15 to 17 weeks after SAH. Of the total samples (37) obtained within 17 weeks after SAH, 8.1% (three of 37) were false negative. No iron-positive cells were detected in samples obtained later than 21 weeks after the SAH episode (10 samples). KEY WORDS 9 cerebrospinal fluid cytology 9 ceil-collecting device subarachnoid hemorrhage 9 ferrous iron staining
T
HE presence of subarachnoid hemorrhage (SAH) is diagnostically confirmed by the detection of bloody and/or xanthochromic cerebrospinal fluid (CSF). However, grossly, CSF returns to its normal clear appearance within 3 weeks after SAH; therefore, after a lapse of time, macroscopic CSF examination may not be useful in confirming the previous occurrence of SAH. Using a device for collecting CSF cells that we have described earlier, x we discovered a diagnostic method that facilitates the confirmation of SAH 1 to 17 weeks after its occurrence. Materials and Methods
We collected 47 lumbar samples from 32 patients with ruptured cerebral aneurysms at various times after the bleeding episode. We excluded from the study SAH cases complicated by either intracerebral or intracranial hematoma, or presenting with multiple SAH's. Ten lumbar samples from non-SAH cases served as the control. Cells from 4-ml CSF samples were collected onto microscopic slides using the sedimentation chamber method described elsewhere, x'2 For each experiment, two sets of slides were prepared. Smear samples were 352
9
either fixed by an alcohol-formalin mixture, stained with Perl's reagent to identify intracellular ferrous iron, and counterstained with Kernechtrot, or, alternatively, stained with May-Gruenwald-Giemsa to identify cellular type. Differential counts of the iron-positive cells were made. In each C S F sample, blood a n d / o r xanthochromia content was estimated and graded from + to + + + + +. The following criteria were used for the gradations: + + + + + = pure blood, + + + + = turbid with red blood corpuscles with or without xanthochromia, + + + = heavy xanthochromia, + + = moderate xanthochromia, and + = slight xanthochromia. Results
The microscopic appearance of iron-positive cells varied greatly, but all were identified as belonging to the mononuclear-phagocytic system 4 (Fig. 1). Cellular morphology was identical in all samples examined, and this finding is in good agreement with that of previous studiesY Grossly, CSF became clear within 3 weeks of the SAH episode (Fig. 2). Iron-positive cells were noted in the CSF at the first week after the onset of SAH. The J. Neurosurg. / Volume 51 / September, 1979
Persistence of iron-positive cells after SAH ratio of iron-positive cells to total nucleated cells rose steeply during the period from 2 to 6 weeks, reached the maximum (8.6%) at 4 to 6 weeks, and decreased gradually to 4.7% at 10 to 12 weeks. Between 15 to 17 weeks, iron-positive cells were noted in 1% of the samples. However, no iron-positive cells were detected later than the 21st week (Fig. 3). All 27 samples obtained between 2 and 9 weeks after the onset of SAH were positive for ironcontaining cells. One of the four samples (25%) obtained during the first week had no iron-positive cells (false negative samples); similarly, one of three samples (33%) obtained between the 10th and 12th weeks and the 15th and 17th weeks were false negative. Therefore, three of 37 (8.1%) of the samples were false negative (Table 1). Discussion
FIG. 1. Photomicrographs of iron-positive cells. The cytoplasm of the macrophage is stained uniformly and/or granularly blue. Perl's reagent and Kernechtrot stain, • 2000.
The present study confirmed earlier findings 5 that on gross examination, CSF becomes clear within 3 weeks of SAH onset. Therefore, it is difficult to obtain an unequivocal diagnosis by routine CSF examination 3 weeks after the occurrence of an SAH episode. Furthermore, the electrophoretic change in CSF protein normalizes within 2 weeks after SAH. 6 However, our present data indicate that iron-positive cells may be detected in samples obtained up to 17 weeks after the onset of SAH. These findings suggest that until the 17th week, an SAH episode can be confirmed. Sayk 5 reported that he detected no iron-positive cells in CSF smears obtained later than 6 weeks after SAH. This discrepancy may be due to the high CSF cell recovery rate (about 80%) facilitated by our cell-collecting device, la Wiezorek 8 attempted cyto-diagnosis of SAH using May-Gruenwald-Giemsa stained smears. We improved on his method by staining intracellular iron with Perl's reagent, which renders the detection of iron-positive macrophages easier and more precise. Occasionally, Perl's reagent stained the cytoplasm of mononuclear phagocytic cells uniformly, and sometimes granular structures, compatible with hemosiderin, were evident. Uniform cytoplasmic staining may be due to a dispersed deposition of iron in ferritin form. Slight but recurrent hemorrhages into the subarachnoid space with occult xanthochromia in CSF
+++ re OIJ-
Z
0
7A ++
/A
/A
"/1
oN 0
m
U_ 0 W t21
0 -I0 0 T t-Z
/A (A /A /A
at .
/`1 /A
/A /A /.4
//1 /`1 //I /`1 /.1
n, X
/A /A /A
t 'TIME 2
SAH
5
6
IN W E E K S
FIG. 2. Graph of the gross appearance of the cerebrospinal fluid sample. Degree of xanthochromia is classified as described in the text. Means of the grade score (mean number of +'s) are shown for each week.
TABLE 1 Total and false-negative samples at different time periods after subarachnoid hemorrhage
Samples no. of samples no. of false-negative samples
Week 1
Week 2
4
5
9
6
7
3
0
3
37
1
0
0
0
0
1
0
l
3
J. Neurosurg. / Volume 51 / September. 1979
W e e k Weeks Weeks Weeks Weeks Weeks 3 4-6 7-9 10-12 13 & 14 15-17
Total
353
U. Ito a n d Y. I n a b a or) ._m _#
,,, I0 (..)
L_
W >
p--
(/) 0 I
Z
o
rr
5
it. O i,i (.9
Cerebrospinal fluid cytology after subarachnoid hemorrhage UMEO ITO, M.D., AND YUTAKA INABA,M.D.
Department of Neurosurgery, Tokyo Medical and Dental University, Tokyo, Japan t,' A method is described which has been found capable of detecting subarachnoid hemorrhage (SAH) up to 15 to 17 weeks after its occurrence. The episode of SAH was confirmed by bloody and/or xanthochromic cerebrospinal fluid (CSF) at the time of SAH onset. In this study, 47 samples of lumbar CSF from diagnostically confirmed SAH patients were used. The CSF cells were collected onto slides and stained with May-Gruenwald-Giemsa or Perl's reagent. Iron-positive cells were detected at I week, increased by 4 to 6 weeks to 8.5% of total nucleated cells, and decreased to 1% by 15 to 17 weeks. All 27 samples obtained at 2 to 9 weeks after SAH showed iron-positive cells. No iron-positive cells (false-negative samples) were noted in 25% (one of four) of samples obtained during the first week, and in 33% (one of three) of samples obtained 10 to 12 weeks and 15 to 17 weeks after SAH. Of the total samples (37) obtained within 17 weeks after SAH, 8.1% (three of 37) were false negative. No iron-positive cells were detected in samples obtained later than 21 weeks after the SAH episode (10 samples). KEY WORDS 9 cerebrospinal fluid cytology 9 ceil-collecting device subarachnoid hemorrhage 9 ferrous iron staining
T
HE presence of subarachnoid hemorrhage (SAH) is diagnostically confirmed by the detection of bloody and/or xanthochromic cerebrospinal fluid (CSF). However, grossly, CSF returns to its normal clear appearance within 3 weeks after SAH; therefore, after a lapse of time, macroscopic CSF examination may not be useful in confirming the previous occurrence of SAH. Using a device for collecting CSF cells that we have described earlier, x we discovered a diagnostic method that facilitates the confirmation of SAH 1 to 17 weeks after its occurrence. Materials and Methods
We collected 47 lumbar samples from 32 patients with ruptured cerebral aneurysms at various times after the bleeding episode. We excluded from the study SAH cases complicated by either intracerebral or intracranial hematoma, or presenting with multiple SAH's. Ten lumbar samples from non-SAH cases served as the control. Cells from 4-ml CSF samples were collected onto microscopic slides using the sedimentation chamber method described elsewhere, x'2 For each experiment, two sets of slides were prepared. Smear samples were 352
9
either fixed by an alcohol-formalin mixture, stained with Perl's reagent to identify intracellular ferrous iron, and counterstained with Kernechtrot, or, alternatively, stained with May-Gruenwald-Giemsa to identify cellular type. Differential counts of the iron-positive cells were made. In each C S F sample, blood a n d / o r xanthochromia content was estimated and graded from + to + + + + +. The following criteria were used for the gradations: + + + + + = pure blood, + + + + = turbid with red blood corpuscles with or without xanthochromia, + + + = heavy xanthochromia, + + = moderate xanthochromia, and + = slight xanthochromia. Results
The microscopic appearance of iron-positive cells varied greatly, but all were identified as belonging to the mononuclear-phagocytic system 4 (Fig. 1). Cellular morphology was identical in all samples examined, and this finding is in good agreement with that of previous studiesY Grossly, CSF became clear within 3 weeks of the SAH episode (Fig. 2). Iron-positive cells were noted in the CSF at the first week after the onset of SAH. The J. Neurosurg. / Volume 51 / September, 1979
Persistence of iron-positive cells after SAH ratio of iron-positive cells to total nucleated cells rose steeply during the period from 2 to 6 weeks, reached the maximum (8.6%) at 4 to 6 weeks, and decreased gradually to 4.7% at 10 to 12 weeks. Between 15 to 17 weeks, iron-positive cells were noted in 1% of the samples. However, no iron-positive cells were detected later than the 21st week (Fig. 3). All 27 samples obtained between 2 and 9 weeks after the onset of SAH were positive for ironcontaining cells. One of the four samples (25%) obtained during the first week had no iron-positive cells (false negative samples); similarly, one of three samples (33%) obtained between the 10th and 12th weeks and the 15th and 17th weeks were false negative. Therefore, three of 37 (8.1%) of the samples were false negative (Table 1). Discussion
FIG. 1. Photomicrographs of iron-positive cells. The cytoplasm of the macrophage is stained uniformly and/or granularly blue. Perl's reagent and Kernechtrot stain, • 2000.
The present study confirmed earlier findings 5 that on gross examination, CSF becomes clear within 3 weeks of SAH onset. Therefore, it is difficult to obtain an unequivocal diagnosis by routine CSF examination 3 weeks after the occurrence of an SAH episode. Furthermore, the electrophoretic change in CSF protein normalizes within 2 weeks after SAH. 6 However, our present data indicate that iron-positive cells may be detected in samples obtained up to 17 weeks after the onset of SAH. These findings suggest that until the 17th week, an SAH episode can be confirmed. Sayk 5 reported that he detected no iron-positive cells in CSF smears obtained later than 6 weeks after SAH. This discrepancy may be due to the high CSF cell recovery rate (about 80%) facilitated by our cell-collecting device, la Wiezorek 8 attempted cyto-diagnosis of SAH using May-Gruenwald-Giemsa stained smears. We improved on his method by staining intracellular iron with Perl's reagent, which renders the detection of iron-positive macrophages easier and more precise. Occasionally, Perl's reagent stained the cytoplasm of mononuclear phagocytic cells uniformly, and sometimes granular structures, compatible with hemosiderin, were evident. Uniform cytoplasmic staining may be due to a dispersed deposition of iron in ferritin form. Slight but recurrent hemorrhages into the subarachnoid space with occult xanthochromia in CSF
+++ re OIJ-
Z
0
7A ++
/A
/A
"/1
oN 0
m
U_ 0 W t21
0 -I0 0 T t-Z
/A (A /A /A
at .
/`1 /A
/A /A /.4
//1 /`1 //I /`1 /.1
n, X
/A /A /A
t 'TIME 2
SAH
5
6
IN W E E K S
FIG. 2. Graph of the gross appearance of the cerebrospinal fluid sample. Degree of xanthochromia is classified as described in the text. Means of the grade score (mean number of +'s) are shown for each week.
TABLE 1 Total and false-negative samples at different time periods after subarachnoid hemorrhage
Samples no. of samples no. of false-negative samples
Week 1
Week 2
4
5
9
6
7
3
0
3
37
1
0
0
0
0
1
0
l
3
J. Neurosurg. / Volume 51 / September. 1979
W e e k Weeks Weeks Weeks Weeks Weeks 3 4-6 7-9 10-12 13 & 14 15-17
Total
353
U. Ito a n d Y. I n a b a or) ._m _#
,,, I0 (..)
L_
W >
p--
(/) 0 I
Z
o
rr
5
it. O i,i (.9
