If you want to minimize x1 under A*x <= b
f = zeros(size(x));
f(1) = 1;
x = linprog(f, A, b);
(x2,...,xn) is not meaningless, it's the associate components such that your constraint A*x <= b is satisfied.
Imagine you want to find most left corner of a 2D triangle.
It can be written as
A*x <= b
with x is 2 x 1, A is 3 x 2, b and 3 x 1.
the LINPROG returns you (x1,x2) coordinates of the left most corner (minimize x1) even if you ignore x2 it's still the ordinate of the corner.