diff --git a/doc/source/whatsnew/v2.1.0.rst b/doc/source/whatsnew/v2.1.0.rst
index 9c64105fb0464..a5520d9e5ca2f 100644
--- a/doc/source/whatsnew/v2.1.0.rst
+++ b/doc/source/whatsnew/v2.1.0.rst
@@ -636,6 +636,7 @@ Reshaping
 ^^^^^^^^^
 - Bug in :func:`concat` coercing to ``object`` dtype when one column has ``pa.null()`` dtype (:issue:`53702`)
 - Bug in :func:`crosstab` when ``dropna=False`` would not keep ``np.nan`` in the result (:issue:`10772`)
+- Bug in :func:`melt` where the ``variable`` column would lose extension dtypes (:issue:`54297`)
 - Bug in :func:`merge_asof` raising ``KeyError`` for extension dtypes (:issue:`52904`)
 - Bug in :func:`merge_asof` raising ``ValueError`` for data backed by read-only ndarrays (:issue:`53513`)
 - Bug in :func:`merge_asof` with ``left_index=True`` or ``right_index=True`` with mismatched index dtypes giving incorrect results in some cases instead of raising ``MergeError`` (:issue:`53870`)
diff --git a/pandas/core/reshape/melt.py b/pandas/core/reshape/melt.py
index cd0dc313a7fb3..74e6a6a28ccb0 100644
--- a/pandas/core/reshape/melt.py
+++ b/pandas/core/reshape/melt.py
@@ -141,8 +141,7 @@ def melt(
     else:
         mdata[value_name] = frame._values.ravel("F")
     for i, col in enumerate(var_name):
-        # asanyarray will keep the columns as an Index
-        mdata[col] = np.asanyarray(frame.columns._get_level_values(i)).repeat(N)
+        mdata[col] = frame.columns._get_level_values(i).repeat(N)
 
     result = frame._constructor(mdata, columns=mcolumns)
 
diff --git a/pandas/tests/reshape/test_melt.py b/pandas/tests/reshape/test_melt.py
index ea9da4e87240b..bf03aa56a1da2 100644
--- a/pandas/tests/reshape/test_melt.py
+++ b/pandas/tests/reshape/test_melt.py
@@ -437,6 +437,26 @@ def test_melt_ea_dtype(self, dtype):
         )
         tm.assert_frame_equal(result, expected)
 
+    def test_melt_ea_columns(self):
+        # GH 54297
+        df = DataFrame(
+            {
+                "A": {0: "a", 1: "b", 2: "c"},
+                "B": {0: 1, 1: 3, 2: 5},
+                "C": {0: 2, 1: 4, 2: 6},
+            }
+        )
+        df.columns = df.columns.astype("string[python]")
+        result = df.melt(id_vars=["A"], value_vars=["B"])
+        expected = DataFrame(
+            {
+                "A": list("abc"),
+                "variable": pd.Series(["B"] * 3, dtype="string[python]"),
+                "value": [1, 3, 5],
+            }
+        )
+        tm.assert_frame_equal(result, expected)
+
 
 class TestLreshape:
     def test_pairs(self):