Retrieve content between custom tags using RegEx

1) If you intend to ONLY use the content of the tag, you can use the ER below, in case it removes everything between <> :

$conteudo = '<:item>Conteúdo</item>';
print_r( preg_replace("/<.*?>/", "", $conteudo) );

Example available at ideone

2) If you intend to use the tag itself and the content, you can use the ER below:

$conteudo = '<:item>Conteúdo</item>';
preg_match_all( '~<.+?>(.+?)<\/.+?>~' , $conteudo , $retorno );
echo $retorno[1][0];

Example available at ideone

Update

Step 1) replace <...> with a | marker Result: |HEADER||MAIN|ITEM||

Step 3) Remove double markers || by simple |
Result: |HEADER|MAIN|ITEM|

Step 4) Break the string in the markers and filter the null values
Result: array( 1 => 'HEADER' , 2 => 'MAIN' , 3 => 'ITEM' )

$string = '<:header>HEADER</header><:main>MAIN<:item>ITEM</item></main>';

// passo 1
$string = preg_replace( '/<.*?>/' , '|' , $string );

// passo 2
$string = preg_replace('/\|+/', '|', $string);

// passo 3
$string = array_filter( explode( '|' , $string ) );

Note that NOT is ideal, it only solves a problem. The way you generate this string is inappropriate. See a demo on ideone

I do not know if this is what you want, but it works:

$html = "<:item>ConteúdoA</item><:valor>ConteúdoB</valor><:tag>ConteúdoC</tag><:teste>ConteúdoD</teste>";
preg_match_all("/<:(.*?)>/", $html, $arrTag);

foreach($arrTag[1] as $tag)
{
echo $tag;
preg_match_all('/<:'.$tag.'>(.+?)<\/'.$tag.'>/sm', $html, $conteudo);
print_r($conteudo);
}