Replace NA values with the values of another column using Siuba in Python

Question

I have a dataframe (df) like this:

  id         users_x users_y

1 2012-04-01       1      NA
2 2012-04-02       2      NA
3 2012-04-03       3      NA
4 2012-04-04       4      NA
5 2012-04-05       5      NA
6 2012-04-06      NA       6
7 2012-04-07      NA       7
8 2012-04-08      NA      NA
9 2012-04-09      NA      NA

I aim to fill the NA with a new variable users which combines information from both. The the ideal output should be something like this.

id         users_x users_y users

1 2012-04-01       1      NA     1
2 2012-04-02       2      NA     2
3 2012-04-03       3      NA     3
4 2012-04-04       4      NA     4
5 2012-04-05       5      NA     5
6 2012-04-06      NA       6     6
7 2012-04-07      NA       7     7
8 2012-04-08      NA      NA    NA
9 2012-04-09      NA      NA    NA

Using Siuba from Python on this fashion df >> mutate(users = if_else(isna(_.users_x), _.users_y, _.users_x)). Yet I am having problems to identify the NA using isna() method (NameError: name 'isna' is not defined). Is there a way to overcome this problem? Thanks