diff --git a/backend/src/middleware/hashtags/extractHashtags.js b/backend/src/middleware/hashtags/extractHashtags.js
index d13511e53..04c0322c6 100644
--- a/backend/src/middleware/hashtags/extractHashtags.js
+++ b/backend/src/middleware/hashtags/extractHashtags.js
@@ -1,17 +1,21 @@
 import cheerio from 'cheerio'
+import {
+  exec,
+  build
+} from 'xregexp/xregexp-all.js'
 // formats of a Hashtag:
 //   https://en.wikipedia.org/w/index.php?title=Hashtag&oldid=905141980#Style
 // here:
 //    0. Search for whole string.
-//    1. Hashtag has only 'a-z', 'A-Z', and '0-9'.
+//    1. Hashtag has only all unicode characters and '0-9'.
 //    2. If it starts with a digit '0-9' than 'a-z', or 'A-Z' has to follow.
-const ID_REGEX = /^\/search\/hashtag\/((\p{L}+[\p{L}0-9]*)|([0-9]+\p{L}+[\p{L}0-9]*))$/g
+const regX = build('^\/search\/hashtag\/((\\pL+[\\pL0-9]*)|([0-9]+\\pL+[\\pL0-9]*))$')
 
 export default function (content) {
   if (!content) return []
   const $ = cheerio.load(content)
   // We can not search for class '.hashtag', because the classes are removed at the 'xss' middleware.
-  //   But we have to know, which Hashtags are removed from the content es well, so we search for the 'a' html-tag.
+  //   But we have to know, which Hashtags are removed from the content as well, so we search for the 'a' html-tag.
   const urls = $('a')
     .map((_, el) => {
       return $(el).attr('href')
@@ -20,7 +24,7 @@ export default function (content) {
   const hashtags = []
   urls.forEach(url => {
     let match
-    while ((match = ID_REGEX.exec(url)) != null) {
+    if ((match = exec(url, regX)) != null) {
       hashtags.push(match[1])
     }
   })