Long-term outcomes after percutaneous cryoablation of abdominal wall endometriosis

The median follow-up time was 40.5 months (IQR 26.5–47.2 months). No patient was lost to follow-up. The median VAS score before cryoablation was 8 (IQR 7–9). At the last follow-up, the median score had reduced to0 (IQR 0–0; p < 0.001). All but one patient (97.5%, 39/40) exhibited reductions in the post-procedure VAS scores at a minimum of 12 months; no patient reported a post-procedure VAS score higher than the pre-cryoablation score. The median pain-free survival rates were 89.2% [95% CI 70.1–96.4%] at 36 months, and 76.8% [95% CI 55.3–83.8%] after 60 months ( Fig. 2 ). Eight patients reported pain (median VAS score 5, IQR 3–5.5, p = 0.0008) during follow-up. Their characteristics after the first cryoablation are summarized in Table 3 . Five presented with immediate residual pain at the 3-month post-procedure consultation, and three with recurrent pain after 12 months. One patient (2.5%, 1/40) underwent successful surgical resection, five (12.5%, 5/40) underwent a second procedure, and two (5%, 2/40) refused further management as they considered the pain reduction satisfactory. Of the five patients who requested repeat cryoablation, two procedures were successful. In one patient, the second cryoablation was unsuccessful, and a third cryoablation was performed (in the absence of recurrence) at 32 months of follow-up. Another patient underwent medical treatment after a second cryoablation, indicated by residual pain and the appearance of a new nodule distant from the treated site. One patient became pregnant, and the second procedure was delayed to a time after birth but has not yet been performed at the time of this writing. The primary cryoablation efficacy rates were 80% (32/40) per patient and 85% (44/52) per nodule. The secondary efficacy rates were 87.5% (35/40) per patient and 90% (47/52) per nodule. Most treated lesions (69%, 36/52) were typical on MR imaging; heterogeneous lesions with microcystic inclusions evident in the T2 hypersignals associated with hemorrhagic spots apparent in the spontaneous T1 hypersignals, and were either well-limited nodules (77%, 40/52) or infiltrative patches (23%, 12/52). Two patients treated before May 2022 refused additional MR imaging after 12 months, of whom one reported persistent pain (patient 8, Table 3 ). Three patients with residual pain and residual nodules exhibited hemorrhagic spots at the edges of the ablation zones of the spontaneous T1 high signal intensity ( Fig. 3 ). In patients with recurrent symptoms, MR imaging detected reconstitution of nodules that were often more prominent on US. When T1-weighted MR images featured contrast injection at the discretion of the radiologist, all post-injection enhancement was nonspecific; and such enhancement was also observed in asymptomatic patients. Such findings were apparent at both months 3 and 12 and presented as annular enhancements indicating mild inflammation. Distant enhancement after month 12 served as an additional indicator of recurrence in patients with recurrent pain. Asymptomatic patients who underwent additional MR imaging later than 12 months after treatment exhibited simple subcutaneous and/or intramuscular fibrous sequelae. Subsequent evaluations, particularly those performed after 5 years, revealed complete restoration of the treated areas in 13 patients; there were no residual scars and no enhancement. Three patients (7.5%, 3/40) experienced adverse events in the days following the procedure. Two patients developed grade 1 events, of which one was a strictly subcutaneous hematoma at the puncture site (as revealed by US) which resolved completely by 1 month. The second event was extensive edema that required no special management. One patient developed a grade 3 adverse event related to severe pain over the treated area in the immediate postoperative period. Management required morphine titration leading to acute urine retention and overnight monitoring. In the long term, no patient presented with procedure-associated transitory or permanent sequelae (nerve damage or an esthetic issue). All entry point scars were invisible at 12 months. No patient presented with a parietal hernia or required reconstructive surgery (such as abdominoplasty). To define factors prognostic of cryoablation failure, univariate analysis was performed on patient and lesion characteristics ( Table 4 ). No significant correlation was found between recurrence of pain and deep endometriosis (Fisher exact test p > 0.9999). Although the median length of the longest lesion axis was slightly greater in those with persistent pain (38 versus 27 mm), the difference did not attain statistical significance ( p = 0.066). No other characteristic differed significantly between the two groups. The frequency of persistent pain was similar in patients with nodules and infiltrative zones ( p = 0.09 and p = 0.072 respectively).

This single-institution, retrospective, observational cohort study was conducted in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of the Institution. The Committee waived the need for informed patient consent. The institutional database was retrospectively searched to identify all patients who had undergone percutaneous AWE cryoablation from January 2013 to May 2022, with a minimum follow-up of 12 months. A total of 40 consecutive patients presenting with 52 AWE nodules were identified and included in the study. No patients were excluded because of a lack of 12-month follow-up data. The median age was 37 years (min 22 years; max 45 years) and 32% (27/40) of patients exhibited deep endometriosis. Four patients had never undergone a cesarean section. One had received a total colectomy to treat ulcerative colitis; another had undergone a right inguinal hernia repair. The third patient had undergone left inguinal lymph node dissection to treat a lymphoma, and the fourth had laparoscopic surgery to treat deep endometriosis. Patient and nodule characteristics are presented in Tables 1 and 2 respectively. AWE diagnosis was confirmed either pathologically (when a biopsy was performed; 10%, 4/40), when a cluster of clinico-radiological assessments (90%, 36/40) revealed an abdominal mass that developed after abdominopelvic surgery (most often cesarean section with the Pfannenstiel incision) and/or was associated with (often cyclic) pain, and when MR imaging suggested AWE [ 16 ]. Our institutional MR imaging protocol (1.5- or 3-Tesla systems) includes T2-weighted axial and sagittal sequences and T1-weighted DIXON axial sequences without gadolinium contrast injection; the radiologist can order contrast-enhanced sequences if desired. The post-injection T1-weighted sequences were obtained after administration of an intravenous bolus of 0.1 mmol/kg (0.5 mmol/mL) gadoteric acid (Dotarem; Guerbet, Roissy, France) at 2 mL/s, followed by a 20-mL saline flush. Of all patients, five had two nodules, two had three nodules, and one patient had four. The median major axis of all treated lesions was 29.5 mm (min 10 mm; max 72 mm) in length and all lesions were entirely intramuscular (within the rectus abdominis muscle, 44%; subcutaneous tissue, 40%; and mixed, 16%). Only four patients (10%) underwent biopsies to confirm endometriosis. Cryoablation was performed percutaneously with patients either under general anesthesia (20%, 8/40) or under sedation combined with local anesthesia (80%, 32/40). All procedures were performed under sterile conditions by one of five interventional radiologists with 5–30 years of experience. US and/or computed tomography (CT) guidance was employed depending on nodule size and location ( Fig. 1 ). One to four sterile 17-G cryoprobes (IceRod or IceSeed; Galil Medical, Yokneam, Israel) were used to treat each nodule, with larger nodules requiring more cryoprobes. The median value of the number of cryoprobes used per nodule was one probe. The cryoprobes were inserted along the long axes of the masses, parallel to the skin, as previously published in AWE, adenomyosis, and desmoid tumors or vascular malformations [ 4 , 10 , 11 , 17 – 19 ]. No limit was imposed on the number of nodules that might be treated in a single session. Two successive freezing cycles using argon gas were delivered, up to 10 min each, to each nodule, separated by the same duration of passive heating. Real-time imaging with US or CT was used to monitor progress. The cycle duration and power were optimized to the size of each target nodule; care was taken to treat the entire nodule. Extension of the ice to the skin was meticulously avoided during all cycles. Carbodissection (injection of CO 2 ) or hydrodissection (injection of 5% dextrose) was performed when treating 30/40 patients (75%) to push back at-risk structures and to limit the risk of thermal damage. Once ice had covered the entire targeted lesion, the probes were removed via passive or active thawing. Oral analgesics that included acetaminophen (Perfalgan 1 g; Bristol-Myers Squibb, Rueil-Malmaison, France) and prednisone (Cortancyl; Sanofi, Paris, France) were initiated immediately after cryoablation for 1 week; the doses varied by the pain experienced by the patients. Of the 40 patients, 90% (36/40) were treated on an outpatient basis, but four (10%, 4/40) were admitted as three lived far from the hospital and one required postoperative pain management. The primary endpoint was the long-term clinical efficacy of cryoablation, defined as the percentage of patients completely free of pain at least 12 months after the procedure. AWE-associated pain was systematically assessed using the VAS and scored at the pre-procedural consultation, then again at the 3-, 6-, and 12-month follow-up consultations, and annually thereafter. Residual pain was defined as persistence of symptomatology at the treated site, thus a VAS score > 0 after the 3-month follow-up consultation. Recurrence was defined as pain (VAS score > 0) at the treated site after a pain-free interval noted during the consultations. Deep pelvic endometriosis pain was distinguishable from parietal pain; the latter was perceived as more intense and superficial, and featured a palpable trigger point and nodule. In patients with deep endometriosis, only nodule- or procedure-related parietal pain was assessed. One secondary objective was to review the MRI/US data (when available) to evaluate whether nodule involution correlated with the symptoms. Patients treated prior to May 2022 received either a face-to-face or a telephone consultation during which follow-up imaging was proposed (especially if the VAS score was > 0) to evaluate the presence of any residual tissue. The appearance of a new nodule distant from the treated site was not considered to be a recurrence, nor was an MRI scan that raised a suspicion of recurrence if the patient had no symptoms. Another secondary objective was to review the procedural adverse events and any side effects recorded during follow-up. These were graded using the classification of the Cardiovascular and Interventional Radiological Society of Europe [ 20 ]. Grades 1 and 2 indicate “minor” complications and grades 3 through 6 indicate “major” complications. Patient characteristics were retrieved from the medical records and the body mass index (BMI) calculated as weight (kg)/height (m) 2 . Deep endometriosis was diagnosed if the condition was revealed by US and/or MRI; the latter was used to define nodule location (subcutaneous, intramuscular, or mixed) and size. The largest nodule diameter was that defined by the RECIST (1.1) criteria [ 21 ]. Nodule volume was calculated using the ellipsoid formula that employs the three largest diameters apparent on T1- or T2-weighted MRI images (volume = length × width × height × 0.5233). The inclusion criteria were age over 18 years and at least one symptomatic AWE for which cryoablation was technically feasible. Patients who did not undergo at least one MRI session followed by a consultation at 12 months post-procedure were excluded. In this descriptive study, qualitative data are expressed as proportions with percentages, and quantitative data as medians with IQRs. All data were compiled and analyzed, and all images retrospectively reviewed, by two authors working independently. One was a radiologist-in-training and the other a radiologist with over 10 years of experience. The primary efficacy rate was the percentage of subjective pain relief after the initial procedure. The secondary efficacy rate was defined as a need for at least one additional cryoablation procedure to treat residual pain or pain that recurred in the treated area. Patient and nodule characteristics were subjected to univariate analysis by recurrence status after the first cryoablation. The Fisher exact test was used to compare qualitative variables and the Mann–Whitney U -test to compare quantitative variables; a p -value < 0.05 was considered statistically significant [ 22 ]. Survival curves were constructed employing the Kaplan–Meier estimator. All statistical analyses were performed using stats.pvalue.io (Paris, France) online biostatistics software. Data were entered into worksheets for storage and extraction/analysis (Excel; Microsoft, Redmond, WA, USA).

The results of this long-term study show that percutaneous cryoablation is a safe and effective, and a minimally invasive option when treating AWE. Few adverse events were reported during follow-up and there were no long-term esthetic sequelae. Real-time imaging of ice balls during cryoablation eliminated any risk of skin burns and peritoneal damage. We added accessory techniques such as superficial thermal protection, hydrodissection, and/or carbodissection [ 23 – 25 ], as needed. The use of local anesthesia/sedation, the mild post-cryoablation symptoms, and the lack of any need for a complex dressing accelerated recovery and reduced the hospital stay. We co-operated with our gynecology department (which operates an outpatient clinic); almost all AWE patients required outpatient care only [ 17 , 26 ]. Cryoablation has consistently achieved local AWE control in the short, medium-, and long-term [ 4 , 10 , 11 ]. The outcomes reported in this long-term study are similar to those afforded by the surgical or other noninvasive treatment options, including the high-intensity focused US technique previously developed to treat AWE [ 27 , 28 ]. Although that method is associated with low recurrence rates (up to 8%) and only minor complications [ 27 , 29 ], cryoablation may be preferable to heat-based ablation because it is possible to monitor ice build-up in real-time via imaging, ice damages the collagen matrix less than does heat, and ice is associated with less tissue granulation than heat ultimately enhancing the long-term esthetic results [ 30 ]. However, some patients reported transient inflammation and discomfort post-cryoablation. A careful pre-procedural consultation must explain the possibility of such symptoms, and adequate post-procedural management is essential [ 31 , 32 ]. The primary and secondary efficacy rates in terms of symptom relief were 80% and 87.5%, respectively. Symptom management in patients with endometriosis remains challenging, and cryoablation is appropriate for only a subset of patients [ 33 ]. The prevalence of deep endometriosis was higher (32%) in our population than in a general endometriosis population (10%) [ 34 ]. This may reflect selection bias; patients with more severe disease may be more frequently referred by gynecologists for treatment after screening [ 35 , 36 ]. However, the recurrence rates after surgical management of endometriosis nodules ranged from 4.3 to 11.4% [ 33 , 37 , 38 ]. Given the benign nature of AWE and thus the need to reduce the risks of complications [ 39 ], concerns about “adequate margins” are less pressing than when malignant tumors or more aggressive lesions (such as desmoids) are encountered [ 18 ]. Nevertheless, further attention to treatment margins may yield even better results; MRI guidance might be valuable in this context. Further work should explore whether cryoablation is of assistance in the very early stages of endometriosis, when nodules lie deep, and whether the addition of hormonal therapies would improve outcomes even further. This retrospective non-comparative study had certain inherent limitations, particularly the small number of patients (because of both the rarity and the likely underestimation of AWE) [ 40 ]. Case-by-case discussions during multidisciplinary team meetings may introduce selection bias. The lack of any control group means that we cannot definitively claim that cryoablation is superior to surgical excision or medical treatment [ 41 ]. The incidence of treatment failure was low and it was therefore not possible to identify predictive factors. Not all patients underwent biopsies; these are essential when the clinical and MRI findings do not reveal characteristic features of AWE, especially when lesions do not lie on cesarean section scars [ 16 ]. Some patients had undergone pre-procedural biopsies to exclude other potential diagnoses, such as desmoid tumors or endometrial cancer [ 42 ]. In summary, cryoablation is a minimally invasive procedure that, in the long-term, safely and effectively treats AWE. No patient or lesion characteristic predictive of cryoablation failure was identified. Future prospective multicenter clinical trials are warranted to evaluate cryoablation further. The procedure should perhaps be considered for incorporation into the recognized AWE treatment recommendations.

Abdominal wall endometriosis (AWE) refers to the ectopic implantation of endometrial tissue within abdominal structures that lie superficial to the peritoneum [ 1 ]. The incidence of AWE ranges from 0.03 to 3.5% with an average age at diagnosis of 35 years [ 2 ]. AWE is associated with previous cesarean section, with 57% of cases occurring in women who have undergone cesarean section, [ 3 ] but spontaneous parietal nodules in women lacking of any surgical history have been described in the umbilicus or the inguinal region as well [ 4 – 7 ]. The predominant symptom of AWE is abdominal pain, which compromises quality-of-life [ 8 ]. Ultrasound (US) and magnetic resonance (MR) imaging are essential for diagnosis, both to rule out other possible diagnoses and to accurately map the lesions [ 9 ]. When imaging is equivocal, a percutaneous biopsy may be offered [ 1 ]. Treatment of symptomatic AWE traditionally employed hormonal therapy and/or surgery [ 1 ]. Percutaneous cryoablation has been developed over the last decade, and has been offered when medical management failed or was refused [ 10 – 13 ]. Initial short-term results revealed remarkable safety and efficacy profiles [ 11 – 15 ]. Recent medium-term reports described significant pain reduction, from a median pain score of 8/10 (interquartile range [IQR] 7–9) on the visual analog scale (VAS) to 0/10 (IQR 0–1) at the last follow-up ( p < 0.0001), with median pain-free survival rates of 93.75% [95% confidence interval [CI] 77.25–98.4%] at 6 months and 82.72% [95% CI 58.8–93.45%] at 12 months respectively [ 4 ]. The aim of this retrospective study was to assess the long-term outcomes after cryoablation of AWE.