BUG: Handling columns from index_col in _is_potential_multi_index

pandas/io/parsers.py

@@ -1168,7 +1168,7 @@ def _is_index_col(col):
     return col is not None and col is not False
 
 
-def _is_potential_multi_index(columns):
+def _is_potential_multi_index(columns, index_col=None):
     """
     Check whether or not the `columns` parameter
     could be converted into a MultiIndex.
@@ -1177,15 +1177,20 @@ def _is_potential_multi_index(columns):
     ----------
     columns : array-like
         Object which may or may not be convertible into a MultiIndex
+    index_col : None, bool or list, optional
+        Column or columns to use as the (possibly hierarchical) index
 
     Returns
     -------
     boolean : Whether or not columns could become a MultiIndex
     """
+    if index_col is None or isinstance(index_col, bool):
+        index_col = []
+
     return (
         len(columns)
         and not isinstance(columns, MultiIndex)
-        and all(isinstance(c, tuple) for c in columns)
+        and all(isinstance(c, tuple) for c in columns if c not in list(index_col))
     )
 
 
@@ -1570,7 +1575,7 @@ def _maybe_dedup_names(self, names):
         if self.mangle_dupe_cols:
             names = list(names)  # so we can index
             counts = defaultdict(int)
-            is_potential_mi = _is_potential_multi_index(names)
+            is_potential_mi = _is_potential_multi_index(names, self.index_col)
 
             for i, col in enumerate(names):
                 cur_count = counts[col]

pandas/tests/io/excel/test_readers.py

@@ -1084,3 +1084,7 @@ def test_excel_high_surrogate(self, engine):
         # should not produce a segmentation violation
         actual = pd.read_excel("high_surrogate.xlsx")
         tm.assert_frame_equal(expected, actual)
+
+    @pytest.mark.parametrize("filename", ["df_empty.xlsx", "df_equals.xlsx"])
+    def test_header_empty_cells(self, engine, filename):
+        pd.read_excel(filename, sheet_name="Sheet1", index_col=0, header=[0, 1])