Letters to the editor |
Reply: Routine use of hysterosalpingography prior to diagnostic laparoscopy in the fertility workup
1 Department of Gynaecology, Division of Reproductive Medicine and Clinical Epidemiology, Leiden University Medical Center, RC Leiden 2 Department of Obstetrics and Gynaecology, Medical Center Haaglanden, The Hague, The Netherlands and 3 Department of Gerontology & Geriatrics, Leiden University Medical Center, RC Leiden, The Netherlands
4 To whom correspondence should be addressed at: Department of Gynaecology, Division of Reproductive Medicine, Leiden University Medical Center, P.O. Box 9600, 2300 RC Leiden, The Netherlands. E-mail: f.m.helmerhorst{at}lumc.nl
Sir,
We appreciate the correspondents’ interest in our study. They hold that in diagnostic trials ‘...randomized comparisons of tests can only be validly evaluated in terms of patient outcomes in case an abnormal test result is followed by a standardized and explicit treatment protocol’. As stated, first we ensured that hysterosalpingography (HSG) and laparoscopy results were uniformly interpreted in all three hospitals after HSG. Following the test results, the participating hospitals used their own treatment protocols for therapeutic reproductive surgery and assisted reproductive treatments. The interventions that followed an abnormal diagnostic laparoscopy are well-accepted in clinical practice, as surgery or IVF for tubal pathology, surgery and/or IVF for endometriosis. A standardized and explicit treatment protocol would have altered the practice setting in which we wished to conduct our trial. If we had imposed on this trial some treatment algorithm, that would have confined the generalizability of the trial to centres with an identical algorithm and would have prevented answering the question whether there is use for HSG in routine practice settings.
The authors also express their concern about the power of the study. They state that ‘...a difference in outcome could only come from patients with discordant test results...’. Furthermore, they estimate that ’between 79 and 97% of the patients’ would not contribute to a difference in pregnancy rate. If indeed the majority of patients undergoing HSG would not benefit, should that not raise questions about the use of HSG as routine policy?
The correspondents suggest a change in design of the study by only randomizing patients with discordant test results. However, the suggested change in design would have resulted in an efficacy trial and would therewith have evaluated the effect of the test under ideal circumstances. In such a trial the question ‘can it work?’ is answered (Haynes, 1999
). However, as stated, our trial was an effectiveness trial designed to compare two common clinical policies in the setting of daily clinical practice. Hence, in our trial the question ‘does it work in practice?’ was answered (Haynes, 1999). In that setting, a worthwhile test plus intervention must have an effect that is robust enough to overcome differences in patient choice, local procedures and health care access.
In conclusion, our study was a pragmatic effectiveness trial designed to evaluate whether two commonly used diagnostic strategies have a different impact on pregnancy rates. Until others design a complete efficacy trial, the evidence from this trial indicates that there is little to choose between the two diagnostic strategies.
Reference
Haynes B. (1999) Can it work? Does it work? Is it worth it? BMJ 319:652–653.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||