SQL Server uses a B+-Tree index not the traditional B-Tree index. There are two major differences between traditiona B-Tree index and B+-Tree index:
- In a B+-Tree index, all of the data is in the leaf level pages. Only key values and pointers exist in the root and intermediate levels.
- In a B+-Tree index, there are pointers which point to the next and last page in the level.
With each page pointing to the page ahead and behind you can do range scans without having to travel up and down the index levels.
On Wikipedia here, you will find SQL Server is one of many modern database engines using B+, a reference has been cited in the page.
When you read Microsoft documentation here, you know by the description how indexes work on SQL Server that B+ is used and not the traditional B-tree.