Christiaans et al evaluated clinical findings that might help determine if a patient with a history of cancer has intracranial metastases. The authors are from 4 hospitals in The Netherlands.
Parameters:
(1) duration of new headache or change in nature
(2) type of headache
(3) emesis
Parameter |
Finding |
Odds Ratio |
95% CI |
p value |
duration |
<= 10 weeks |
11.0 |
1.1 to 108.2 |
0.04 |
type |
not tension type |
6.7 |
1.8 to 25.1 |
0.004 |
emesis |
present |
4.0 |
1.1 to 14.3 |
0.03 |
Parameter |
Finding |
Points |
duration of headache |
<= 10 weeks |
2.4 |
|
> 10 weeks |
0 |
type of headache |
tension type |
0 |
|
not tension type |
1.9 |
emesis |
absent |
0 |
|
present |
1.4 |
where:
• The points assigned are equal to LN(odds ratio) for each parameter.
index =
= SUM(points for all 3 parameters) – 3.9
Interpretation:
• minimum score: -3.9 (absence of all 3 findings)
• maximum score: 1.8
Index |
Situations |
Risk of Intracerebral Metastases |
-3.9 |
none of the 3 findings |
very low |
-2.5 to –2.0 |
vomiting only, headache type only |
low |
-1.5 to –0.1 |
duration only, vomiting and headache type, vomiting and duration |
intermediate |
0.4 to 1.8 |
headache type and duration, all 3 findings |
high |
after Table 3, page 2068
where:
• The value for emesis + headache in Table 3 is given as –1.0 but should be –0.1.
Performance:
• The area under the ROC curve was 0.83.
• Because of the low specificity of the findings, it was felt that few patients would not need an MRI to exclude metastases. It was felt that an MRI is warranted in all cancer patients with a new or changed headache of recent onset.
• MRI can be excluded in those patients without any of the 3 symptoms (and this group would not be covered by the headache of new onset or recent change).
Limitations:
• The authors felt the findings need to be evaluated in a larger prospective study.
• The authors mentioned that stage of the disease might be an additional independent risk factor.
• I would think that including frequency data of brain metastases for the patient's tumor might be helpful (i.e, is the tumor one that frequently or rarely goes to brain).
Purpose: To evaluate a cancer patient with headache for possible intracranial metastases usng the clinical criteria of Christiaans et al.
Specialty: Hematology Oncology, Surgery, general
Objective: clinical diagnosis, including family history for genetics
ICD-10: C79.3,