diff --git a/pandas/core/nanops.py b/pandas/core/nanops.py
index 6dc05c23c026f..cfb02e5b1e987 100644
--- a/pandas/core/nanops.py
+++ b/pandas/core/nanops.py
@@ -8,7 +8,7 @@
 
 from pandas._config import get_option
 
-from pandas._libs import NaT, Timedelta, Timestamp, iNaT, lib
+from pandas._libs import NaT, Timedelta, iNaT, lib
 from pandas._typing import ArrayLike, Dtype, DtypeObj, F, Scalar
 from pandas.compat._optional import import_optional_dependency
 
@@ -330,7 +330,7 @@ def _na_ok_dtype(dtype: DtypeObj) -> bool:
     return not issubclass(dtype.type, np.integer)
 
 
-def _wrap_results(result, dtype: DtypeObj, fill_value=None):
+def _wrap_results(result, dtype: np.dtype, fill_value=None):
     """ wrap our results if needed """
     if result is NaT:
         pass
@@ -340,15 +340,11 @@ def _wrap_results(result, dtype: DtypeObj, fill_value=None):
             # GH#24293
             fill_value = iNaT
         if not isinstance(result, np.ndarray):
-            tz = getattr(dtype, "tz", None)
             assert not isna(fill_value), "Expected non-null fill_value"
             if result == fill_value:
                 result = np.nan
 
-            if tz is not None:
-                # we get here e.g. via nanmean when we call it on a DTA[tz]
-                result = Timestamp(result, tz=tz)
-            elif isna(result):
+            if isna(result):
                 result = np.datetime64("NaT", "ns")
             else:
                 result = np.int64(result).view("datetime64[ns]")
diff --git a/pandas/tests/test_nanops.py b/pandas/tests/test_nanops.py
index 458fba2e13b0f..359a7eecf6f7b 100644
--- a/pandas/tests/test_nanops.py
+++ b/pandas/tests/test_nanops.py
@@ -988,11 +988,10 @@ def prng(self):
 
 
 class TestDatetime64NaNOps:
-    @pytest.mark.parametrize("tz", [None, "UTC"])
     # Enabling mean changes the behavior of DataFrame.mean
     # See https://github.com/pandas-dev/pandas/issues/24752
-    def test_nanmean(self, tz):
-        dti = pd.date_range("2016-01-01", periods=3, tz=tz)
+    def test_nanmean(self):
+        dti = pd.date_range("2016-01-01", periods=3)
         expected = dti[1]
 
         for obj in [dti, DatetimeArray(dti), Series(dti)]: