This poster was originally presented at the RANZCR Annual Scientific Meeting 2011, October 6-9, in Melbourne/AU.
Congress:
RANZCR ASM 2011
Keywords:
Interventional vascular, Neuroradiology brain, Catheter arteriography, CT, Aneurysms, Haemorrhage
Authors:
A. Lane, N. Stewart, L. Jones, A. Coulthard; Brisbane/AU
DOI:
10.1594/ranzcr2011/R-0149
Results
General numbers:
- 112 patients with aneurysmal SAH treated with coiling were included
- 29 (25.9%) developed angiographic vasospasm
- 23 (20.5%) were treated with IA verapamil or angioplasty
- Age and sex had no relationship to outcome in this study (41 males,
71 females,
age range 15 to 82)
Fisher score:
- A trend is evident but this is not sufficient for an association (see Table 3)
- Patients with high Fisher scores (4) had an odds ratio for developing vasospasm of 1.9 (Cl: 0.61,
5.96) compared to patients with low Fisher scores (1-3),
however this did not prove to be significant (p=0.27)
WFNS score:
- No relationship was demonstrated between WFNS score and risk of developing vasospasm requiring endovascular therapy (see Table 4).
Summed Fisher and WFNS score:
- Patients with summed Fisher and WFNS score > 6 had an odds ratio for vasospasm requiring endovascular treatment of 2.6 (Cl: 0.96,
7.34) compared to patients with lower scores however,
in this small cohort,
this relationship did not prove to be statistically significant (p=0.06) (see Table 5)
Table 3
Fisher score
|
Number of patients
|
Number of patients requiring endovascular treatment for vasospasm
|
1
|
10 (8.9%)
|
0
|
2
|
5 (4.4%)
|
0
|
3
|
25 (22.3%)
|
5 (20%)
|
4
|
72 (64%)
|
18 (25%)
|
Table 4
WFNS score
|
Number of patients
|
Number of patients requiring endovascular treatment for vasospasm
|
1
|
45 (40%)
|
6 (13.3%)
|
2
|
25 (22.3%)
|
5 (20%)
|
3
|
6 (5%)
|
2 (33.3%)
|
4
|
21 (18.8%)
|
5 (23.8%)
|
5
|
15 (13.4%)
|
5 (33.3%)
|
Table 5
Summed Fisher and WFNS score
|
Number of patients
|
Number of patients requiring endovascular treatment for vasospasm
|
Low (1-5)
|
49 (43.8%)
|
6 (12.2%)
|
High (6+)
|
63 (56.2%)
|
17 (27%)
|
Discussion:
We found,
in this retrospective cohort study:
- a trend between Fisher score and risk of development of vasospasm after aSAH,
although the numbers were not sufficient for an association
- WFNS score alone did not have an association with risk of development of vasospasm
- an association was demonstrated when WFNS and Fisher score were summed,
although this was not a statistically significant relationship in this small cohort
These findings are consistent with the current evidence in the field. Fisher score was developed as a grading system to predict vasospasm. However,
WFNS score was developed with a different purpose in mind. In keeping with this,
there have been fewer studies performed to evaluate the relationship between WFNS score and vasospasm. Inagawa et al (15) and Dupont et al (14) did find WFNS grade to be predictive of both symptomatic and angiographic vasospasm (15). However,
Charpentier et al (16) found the opposite: that good neurological grade,
as measured by WFNS score,
was associated with increased risk of cerebral vasospasm.
Limitations:
1. Fisher score 1
- Ten patients were allocated a Fisher score of 1 (see Table 6)
- In the case of seven of these patients,
the presentation CT scan was delayed by a number of days after the timing of the likely bleed,
as assessed by symptom onset
- CT scanning has been reported as having a sensitivity of 100% from day 1 to day 5 for evaluation of suspected SAH(17) however,
after day 5 the sensitivity reduces
- For those patients that had a delayed CT scan in this study,
their Fisher score may have been higher if an earlier scan had been performed
- None of those patients developed vasospasm so the effect of this potential error is unlikely to be significant.
Table 6: Patients with Fisher 1 score
Patient number
|
Delay to initial CT head scan (days)
|
Mode of diagnosis of SAH: LP (LP) vs symptoms plus angiographic evidence of likely ruptured aneurysm (symptoms+angio)
|
9
|
No delay
|
Symptoms+angio
|
12
|
No delay
|
LP
|
45
|
6
|
LP
|
61
|
5
|
Symptoms+angio
|
62
|
No delay
|
Symptoms+angio
|
67
|
7
|
Symptoms+angio
|
78
|
9
|
LP
|
81
|
5
|
Symptoms+angio
|
98
|
16
|
Symptoms+angio
|
99
|
30
|
Symptoms+angio
|
2. Statistical issues
- No patients with a Fisher score of 1 or 2 developed the outcome of interest therefore we were unable to perform a logistic regression unless we categorized the variables
- This forced a categorization of a low (Fisher score 1-3) vs high (Fisher score 4) group
- This grouping is not ideal from a clinical point of view as a Fisher score of 3 is clinically very different to a Fisher score of 1 (and more likely to be thought of as similar to a Fisher score of 4)
3. The small cohort size and the retrospective nature of the study both have innate limitations.