SQL problem: the previous or next record of the record with the primary key ID and sorting condition is known.

Set the record ID to @ ID

The above record is:
Select top 1 * From Table1 where sorting field <(select
Sort field from Table1 where id = @ ID) order by sort field DESC
The following record is:
Select top 1 *
From Table1 where sorting field> (select sorting field from Table1 where id = @ ID) order by sorting Field

======================================
What does it mean to not use a record set? Is a result not a result set, but a value to a variable?

With a single record set, won't you store it into variables? The format is as follows:
The above record is:
Set @ idpre = (
Select top 1
ID from Table1 where sorting field <(select sorting field from Table1 where id = @ ID) order
Sort field DESC
)
The ID of the next record is:
Set @ idpre = (
Select top 1 ID from Table1
Where sorting field> (select sorting field from Table1 where id = @ ID) order by sorting Field
)

======================================
Good question!
Set the sorting fields to field1 int and field2 in sequence.
Varchar (30) field3 datetime
Set @ idpre = (
Select top 1 ID from Table1

Where
(Convert (varchar (10000000000 + field1) + space (30-len (field2) + field2, convert (varchar, field3, 120 ))

>
(Select
Sort field = (convert (varchar (10000000000 + field1) + space (30-len (field2) + field2, convert (varchar, field3, 120 ))

From Table1 where id = @ ID)
Order by sorting Field

-- The key is to convert all the sorted fields into varchar of a fixed length and then combine them together.
======================================

The requirement is not simple. Of course, it is no more simple. In fact, this statement is not very long and the logic is not very complex.