diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index 48a5596e00061..1396aa3690397 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -2957,8 +2957,8 @@ def to_parquet(
           categories, not just those present in the data. This behavior
           is expected and consistent with pandas' handling of categorical data.
           To manage file size and ensure a more predictable roundtrip process,
-          consider using :meth:`Categorical.remove_unused_categories` on the
-          DataFrame before saving.
+          consider using :meth:`Categorical.remove_unused_categories`
+        * When using the "pyarrow" engine, `DataFrame.attrs` are stored as part of the file's metadata and restored on reading.
 
         Examples
         --------