There are a few issues here.
(1) PDF is a “final form file format” and is not a layout format like Word, InDesign, etc. Unless “tagged PDF” was originally created, the PDF file itself does not have a concept of anything other than runs of text, vector objects, and raster image objects, etc. — there are no words, sentences, paragraphs, columns, articles, tables, etc. in PDF (other than if tagging is used, and even then, the PDF file doesn't have a real full view of the original content)! Creation of a Word document (or for that matter an Excel spreadsheet or a Powerpoint presentation) from a PDF file depends on heuristics, hints from any tagging, the quality of the PDF file, and the source of the PDF file. You cannot count on even round-tripping (i.e., Word=>PDF=>Word) and remaining particularly faithful to the original document.
(2) Depending upon the source of the original PDF file, the format you are exporting to in Acrobat (Word, Excel, Powerpoint, etc.) simply might not be able to accommodate the layout, graphic constructs, color spaces, etc. For example, if the original document was a complex InDesign document, you would expect significant “issues” with exported Word documents from the InDesign's PDF simply because many of the layout features in InDesign simply don't exist in Word.
(3) You may wish to check the Word export settings to see whether adjusting those may help in terms of the resultant Word layout.
Unfortunately, the best you should really count on for any export from PDF is an approximation of the original document. Typically, much manual editing of the resultant file will be necessary before it is really usable. You are always best to try to find, edit, and use the original source documents, not count on conversions from PDF.