itextsharp fails to completely parse HTML ordered list

Question

I'm using itextsharp 5.3.2. I've created a custom action using VS2010 to generate a pdf from a SharePoint list item. The string "value" is from an Enhanced Rich Text multiline text field. For the most part, the method below works great to create a PdfPCell with the html content which I can then add to a table in my document. I used the answer from this post to originally write my method: How to get ordered list into pdf using itext?

I've run into a strange issue with an html list, ordered or unordered. It seems that if the list is 3+ levels deep and has multiple items at each level, the html list will fail to parse after the first level 1 item. However, any more content after the ordered list is correctly processed and added to the pdf.

I've wrapped the contents of my method in a try-catch but no exception is caught.

Here is the HTML source of my enhanced rich text field:

<div class="ExternalClassAF9636A4B56F497BB4B962AE682E5CA1"><ol><li>test item level 1</li>
<ol><li>test subitem level 1.1</li>
<ol><li>test subitem level 1.1.1</li>
<li>test subitem level 1.1.2</li>
<li>test subitem level 1.1.3</li></ol>
<li>test subitem level 1.2 <ol><li>test subitem level 1.2.1</li>
<li>test subitem level 1.2.2</li>
<li>test subitem level 1.2.3</li></ol></li>
<li>test subitem level 1.3 <ol><li>test subitem level 1.3.1</li>
<li>test subitem level 1.3.2</li>
<li>test subitem level 1.3.3</li></ol></li></ol>
<li>test item level 2 <ol><li>test subitem level 2.1<ol><li>test subitem level 2.1.1</li>
<li>test subitem level 2.1.2</li>
<li>test subitem level 2.1.3</li></ol></li>
<li>test subitem level 2.2<ol><li>test subitem level 2.2.1</li>
<li>test subitem level 2.2.2</li>
<li>test subitem level 2.2.3</li></ol></li>
<li>test subitem level 2.3<ol><li>test subitem level 2.3.1</li>
<li>test subitem level 2.3.2</li>
<li>test subitem level 2.3.3</li></ol></li></ol></li>
<li>test item level 3 <ol><li>test subitem level 3.1<ol><li>test subitem level 3.1.1</li>
<li>test subitem level 3.1.2</li>
<li>test subitem level 3.1.3</li></ol></li>
<li>test subitem level 3.2<ol><li>test subitem level 3.2.1</li>
<li>test subitem level 3.2.2</li>
<li>test subitem level 3.2.3</li></ol></li>
<li>test subitem level 3.3<ol><li>test subitem level 3.3.1</li>
<li>test subitem level 3.3.2</li>
<li>test subitem level 3.3.3</li></ol></li></ol></li></ol></div>
<p>Some more content after the ordered list.</p>

Here is the method that I'm passing the value into to return a PdfPCell:

const float INDENT = 25f;
...

private static PdfPCell createHtmlCell(string value)
{
    PdfPCell cell = new PdfPCell();  
    cell.BorderWidth = 0f;  
    cell.PaddingLeft = 10f;  

    StyleSheet styles = new StyleSheet();
    styles.LoadTagStyle(HtmlTags.UL, HtmlTags.INDENT, INDENT.ToString());  
    styles.LoadTagStyle(HtmlTags.OL, HtmlTags.INDENT, INDENT.ToString());  
    styles.LoadTagStyle(HtmlTags.LI, HtmlTags.INDENT, INDENT.ToString());  

    using (StringReader sr = new StringReader(value))
    {
        List<IElement> html = HTMLWorker.ParseToList(sr, styles);
        foreach (IElement e in html)
        {
            foreach (Chunk c in e.Chunks)
                c.Font.Size = 11f;

            if (e is Paragraph)
            {
                Paragraph p = (Paragraph)e;
                p.FirstLineIndent = INDENT / 2;
                p.SpacingAfter = 5f;
            }

            cell.AddElement(e);
        }
    }

    return cell;
}

Here is my pdf result: http://screencast.com/t/9OfqZ2c3Vv

Thanks for the help.