Restenosis-narrowing of blood vessels- frequently takes place after coronary anioplasty, but accurate prediction of which individuals will have this problem is problematic. In the study by simons, the authors hypothesized restenosis is more likely to take place if activated smooth muscle cells in coronary lesios at time of surgery are present. They employed the number of reactie nuclei in coronary lesions as an indicator of presence of activated smooth muscle cells. The number of reactive nuclei in lesions and the degree of stenosis at follow up for 16 patients who underwent the second angiography are shown here.
Patient
|
Degreee of Stenosis (%) at Follow-up
|
Number of Reactive Nuclei at initial Surgery
|
1
|
28
|
5
|
2
|
15
|
3
|
3
|
22
|
2
|
4
|
93
|
10
|
5
|
60
|
12
|
6
|
42
|
8
|
7
|
53
|
3
|
8
|
72
|
15
|
9
|
79
|
17
|
10
|
28
|
0
|
11
|
82
|
13
|
12
|
100
|
17
|
13
|
27
|
1
|
Are you suspicious of any of these data points? If so why does there appear to be the linear relation between the degree of stenosis and number of reactive nuclei? If there is explain the relation. Are there any points that have the large influence on the estimated regression line? If there're elminated the point with the greatest leverage and refit equation. Is there much difference between two regression equations? Are there any points that have the large standardized residucal? Describe why the residuals are large for these points. Do you think that simons have a promising lead for the predicting which patients will undergo restenosis?