Your post does not give much details. Do you mean that you first run the stored procedure and it gets stuck on an UPDATE statement. Then you kill the execution of the procedure and then run you the UPDATE statement directly outside the procedure? Or did you first run the procedure from the application, killed the spid, and then ran the procedure directly from SSMS?
In either case, this article on my web site may explain some things for you: http://www.sommarskog.se/query-plan-mysteries.html
[font="Times New Roman"]Erland Sommarskog, SQL Server MVP, www.sommarskog.se[/font]