I don't really understand your question, but if you try to check last update on csv file indexed on splunk, you can use this search to check the date.
| metadata type=sources index=<your_index> | search source=<your_csv_file>
and check date field.
By metadata, I speculate that the "csv file" in question is uploaded for lookup purposes. If so, the easiest solution would be to add a timestamp field to the CSV, and use "| inputlookup" command to check if the timestamp agrees with your latest scheduled update. You can also use a monotonous version field, and compare that field to your upload scheme. Such an auxiliary field does not even have to appear in every row.
what you mean by monotonous version field and how you would set this up?
Same concept as a timestamp field. Let's say in the 1st upload, set a version column to value 1, and each subsequent uploads just increment version, to 2, 3, 4, etc. The operator will need to know which version the latest upload is supposed to be, of course. Therefore, this is less convenient and perhaps less practical than straight timestamp.