The technique described by Rojanasakul is, as first pioneered and reported by Robin Phillips [1] from St. Mark’s Hospital, based on the centrality of chronic intersphincteric anal gland infection in the aetiology and persistence of idiopathic anal fistula, the cryptoglandular hypothesis [2], and the desire for sphincter preservation. Parks actually advocated internal sphincterectomy, i.e. excision of that segment of internal sphincter overlying the intersphincteric space sepsis to ensure adequate drainage, but most surgeons were happy to simply lay it open by internal sphincterotomy as part of an external sphincter preserving strategy [3]. Phillips described an internal and external sphincter conserving surgical approach through the intersphincteric plane with eradication of intersphincteric space sepsis, closure of the internal opening and internal sphincter from within the same plane, and excision of tracks lateral to the plane with closure of the resultant hole in the external sphincter, and primary wound closure. In essence, the only difference from the present description is the treatment of the extrasphincteric component (excision rather than curettage). The 13 fistulas treated in the 1993 series were challenging, in that five were suprasphincteric, three were rectovaginal and five were in patients with inflammatory bowel disease (in which the aetiology may not have resided in cryptoglandular infection, but in which the tract passed across the intersphincteric space); primary healing was achieved in all but one of the patients with an idiopathic trans-sphincteric fistula. One might ask why such a technique has not been more widely adopted. Certainly, there is good supportive evidence that intersphincteric space sepsis is important; its presence is an indicator of a fistula in the acute situation [4], and fistula cure without recourse to sphincter division has always been the fistula surgeon’s ultimate aspiration. An intersphincteric approach, can, for high tracks, be technically demanding; the track must be chronic and well defined by fibrous tissue, and how does one deal with an intersphincteric horseshoe track or one that ascends in the intersphincteric plane to cross voluntary muscle at a higher level than that at which it crossed the internal sphincter? The necessary exposure of the intersphincteric space as shown in figures 5, 6, 7 and 8 must render the internal sphincter susceptible to damage, and a prospective study of structure and function would be welcome. One concept that Rojanasakul fails to mention is the contemporary interest in filling the track with a biomaterial, such as collagen, which might act as a scaffold for host integration. It is unclear whether the range in reported success rates relates to the biomaterial itself or the environment into which the biomaterial is placed (track preparation, addressing secondary extensions, etc.), in addition, of course, to length of follow-up. Coloproctologists are all too aware of the poor levels of evidence in anal fistula surgery, yet we seem content to carry on with our individual preferences based on personal experiences. Surely the time has come for suitably designed and powered prospective randomised studies to answer the questions we, and increasingly, patients ask. This comment refers to the article doi:10.1007/s10151-009-0522-2.