Adenomyosis in endometriosis – prevalence and impact on fertility. Evidence from magnetic resonance imaging
Sir, We read with great interest the paper by Kunz et al. entitled ‘Adenomyosis in endometriosis – prevalence and impact on fertility. Evidence from magnetic resonance imaging’ (Kunz et al., 2005). The authors found a higher incidence of uterine adenomyosis in women with endometriosis than in women without endometriosis and suggested that adenomyosis could be a determinant of infertility (Kunz et al., 2005). However, this article raises issues concerning the magnetic resonance (MR) imaging protocol used to diagnose adenomyosis. Concerning the imaging technique itself, the authors directly adopted a protocol used in a study published five years previously (Kunz et al., 2000) that may not be optimal for the diagnosis of adenomyosis. First, a 1.5-T pelvic phased-array coil with a 256 × 512 matrix offers better spatial resolution than a 1-T body coil with a matrix of 154 × 256, particularly for the detection of hyperintense myometrial spots, which are the findings most specific to adenomyosis. Second, the usefulness of fat-saturated turbo-spin echo sequences for the detection of adenomyosis has never been demonstrated. Third, breath-hold T2-weighted sequences (true fast imaging with steady-state precession and turbo-inversion-recovery sequences) offer better differentiation between focal adenomyosis and uterine contraction, optimize the accuracy of MR imaging for the diagnosis of adenomyosis and reduce interobserver variability, while fast spin-echo T2-weighted images and breath-hold T2-weighted sequences appear to have similar accuracy (Bazot et al., 2003). Concerning the MR imaging criteria, Kunz et al. considered that a junctional zone maximum of >11 mm (JZmax) was alone sufficient for the diagnosis of adenomyosis (Reinhold et al., 1996). In our experience, however, isolated JZmax >11 mm has a sensitivity and specificity of, respectively, 62% and 96% for the diagnosis of adenomyosis (Bazot et al., 2001). The combination of JZ thickness with high-signal-intensity myometrial spots, JZmax/entire myometrium >40% and regular homogeneous uterine enlargement increases the accuracy of MR imaging in women with adenomyosis who do not have associated leiomyomas, raising the sensitivity and specificity to 87% and 100%, respectively (Bazot et al., 2001). Regarding clinical implications, using a JZmax threshold of 10 mm as a criterion of adenomyosis, Kunz et al. found a very high prevalence of adenomyosis in the ‘total endometriotic’ group (79%) compared to both ‘healthy controls’ (9%) and ‘total controls’ (28%) (Kunz et al., 2005). These results contrast with those of a recent study in which only 44 (27%) of 163 women with pelvic endometriosis proven by laparoscopy and histology had adenomyosis on pre-operative MR imaging (Bazot et al., 2004). Finally, like Kunz et al. we also found that uterine adenomyosis was the main determinant of infertility in a series of 34 women undergoing laparoscopic segmental colorectal resection for endometriosis, 22 of whom wished to conceive (Darai et al., 2005). 1Departments of Radiology and 2Obstetrics and Gynecology, Hôpital Tenon, Paris, France