diff --git a/haystack/components/converters/pptx.py b/haystack/components/converters/pptx.py
index c3a32f2079..3f651c7133 100644
--- a/haystack/components/converters/pptx.py
+++ b/haystack/components/converters/pptx.py
@@ -17,12 +17,13 @@
 
 with LazyImport("Run 'pip install python-pptx'") as pptx_import:
     from pptx import Presentation
+    from pptx.oxml.text import CT_RegularTextRun, CT_TextField
     from pptx.shapes.autoshape import Shape
     from pptx.shapes.base import BaseShape
     from pptx.shapes.graphfrm import GraphicFrame
     from pptx.shapes.group import GroupShape
     from pptx.table import Table
-    from pptx.text.text import _Paragraph
+    from pptx.text.text import _Paragraph, _Run
 
 
 logger = logging.getLogger(__name__)
@@ -137,7 +138,13 @@ def _process_paragraph(self, paragraph: "_Paragraph") -> str:
         if self.link_format == "none":
             return paragraph.text
         parts = []
-        for run in paragraph.runs:
+        for child in paragraph._p.iterchildren():
+            if isinstance(child, CT_TextField):
+                parts.append(child.text)
+                continue
+            if not isinstance(child, CT_RegularTextRun):
+                continue
+            run = _Run(child, paragraph)
             if run.hyperlink and run.hyperlink.address:
                 if self.link_format == "markdown":
                     parts.append(f"[{run.text}]({run.hyperlink.address})")
diff --git a/releasenotes/notes/pptx-field-text-30f1b392c307906b.yaml b/releasenotes/notes/pptx-field-text-30f1b392c307906b.yaml
new file mode 100644
index 0000000000..c04867fc20
--- /dev/null
+++ b/releasenotes/notes/pptx-field-text-30f1b392c307906b.yaml
@@ -0,0 +1,7 @@
+---
+fixes:
+  - |
+    ``PPTXToDocument`` now keeps the text of a field, such as a slide number or a date,
+    when ``link_format`` is ``"markdown"`` or ``"plain"``. These formats read only the
+    runs of a paragraph, so the text of a field between runs was dropped, for example
+    the ``3`` in ``Slide 3``. The default ``link_format="none"`` already kept field text.
diff --git a/test/components/converters/test_pptx_to_document.py b/test/components/converters/test_pptx_to_document.py
index cd787f873f..b317d26a63 100644
--- a/test/components/converters/test_pptx_to_document.py
+++ b/test/components/converters/test_pptx_to_document.py
@@ -2,11 +2,16 @@
 #
 # SPDX-License-Identifier: Apache-2.0
 
+import io
 import logging
 import os
 from pathlib import Path
 
 import pytest
+from pptx import Presentation
+from pptx.oxml import parse_xml
+from pptx.oxml.ns import nsdecls
+from pptx.util import Inches
 
 from haystack.components.converters.pptx import PPTXToDocument
 from haystack.dataclasses import ByteStream
@@ -124,6 +129,37 @@ def test_no_link_extraction(self, test_files_path):
         assert "https://example.com" not in content
         assert "Example" in content
 
+    @pytest.mark.parametrize(
+        "link_format,expected",
+        [
+            ("none", "Slide 3 of the deck"),
+            ("markdown", "Slide 3 of [the deck](https://example.com)"),
+            ("plain", "Slide 3 of the deck (https://example.com)"),
+        ],
+    )
+    def test_run_keeps_field_text_with_every_link_format(self, link_format, expected):
+        # A slide-number field (`a:fld`) sits between the runs of the paragraph
+        presentation = Presentation()
+        slide = presentation.slides.add_slide(presentation.slide_layouts[6])
+        paragraph = slide.shapes.add_textbox(Inches(1), Inches(1), Inches(5), Inches(1)).text_frame.paragraphs[0]
+        paragraph.add_run().text = "Slide "
+        paragraph._p.append(
+            parse_xml(
+                f'<a:fld {nsdecls("a")} id="{{B6F15528-21DE-4FAA-801E-634DDDAF4B2B}}" type="slidenum">'
+                "<a:t>3</a:t></a:fld>"
+            )
+        )
+        paragraph.add_run().text = " of "
+        link = paragraph.add_run()
+        link.text = "the deck"
+        link.hyperlink.address = "https://example.com"
+        buffer = io.BytesIO()
+        presentation.save(buffer)
+
+        documents = PPTXToDocument(link_format=link_format).run(sources=[ByteStream(buffer.getvalue())])["documents"]
+
+        assert documents[0].content == expected
+
     def test_run_reads_nested_groups_and_table_with_links(self, test_files_path):
         # The slide includes nested text groups, a grouped table, and an empty decorative shape.
         converter = PPTXToDocument(link_format="markdown")