Resolution of Pain after Percutaneous Image-Guided Cryoablation of Extraperitoneal Endometriosis

This is an observational retrospective study with an approval of the institutional review board. All patients gave their informed consent for the procedure. The informed consent to be included in this study was waived. Retrospective searching in the institutional database identified all patients referred to the IR department between September 2017 and June 2022 for cryoablation of symptomatic EE. The diagnosis of EE was based on the results of pathological analysis or by clinical and radiological features of EE according to Hensen et al.’s criteria [ 20 ]: a mass appeared close to a scar, usually a caesarean scar with Pfannenstiel’s incision, with pain that is often cyclic and predominant in the first phase of the cycle, observed on Doppler ultrasound (US) and magnetic resonance (MR) imaging. Nodule location and size were assessed on preoperative MR imaging and US. Deep pelvic infiltrative endometriosis lesions were considered as untreatable with cryoablation. Deep pelvic infiltrative endometriosis is defined as an extension of endometrial tissue-like beneath the peritoneal surface, typically presenting as nodules and able to invade nearby structures, such as bowel, ureters, and bladder, often accompanied by fibrosis and disturbance of normal anatomy [ 21 , 22 ]. The most common involvement of deep endometriosis concerns the mediocentral compartment of the pelvis, which mainly includes the torus, proximal uterosacral ligaments and posterior vaginal fornix. The presence of associated deep endometriosis was considered when the patient was already being followed for a known deep endometriosis or by the affirmation of the presence of deep endometriosis after review of the preoperative pelvic MR imaging by an expert radiologist. All patient received hormone therapy prior cryoablation (gonadotropin-releasing hormone, oral contraceptives, progesterone, or danazol). Cryoablation was discussed in a multidisciplinary meeting involving surgeons and radiologists when surgery of EE was expected to be too invasive or if the patient refused surgery at the time of the discussion. Inclusion criteria were symptomatic EE and technically feasible cryoablation. No limit in number or location of nodules that were treated at a single session was observed. Patients under 18 years of age were excluded. A total 42 consecutive women were identified from the database and included in this study. Across the 42 patients (median age: 37 years, interquartile range [IQR]: [33 – 39.5]), 47 lesions were treated (2 patients with 2 lesions and 1 patient with 4 lesions). Patient and lesion characteristics are respectively summarized in Table 1 . Twenty-five patients (59.5%, 25/42) had associated deep pelvic endometriosis. EE occurred on a surgical scar in 34 patients (80.9%, 34/42). In the absence of any surgical context, parietal endometriosis appeared in the umbilical (n= 4) and right inguinal (n= 2) regions. The nodule was mainly located in the rectus muscle (36.1%, 17/47) and in the subcutaneous fat (34.0%, 16/47). Median nodule size was 28 mm [17.5 – 30]. Patients were aware that cryoablation was offered as a minimally invasive alternative to surgery and in case of failure assessed during the follow up, a secondary surgical resection may be performed. Patients could be admitted for short hospitalization or as outpatient. The procedures were performed under general anesthesia or local anesthesia combined with sedation and hypnosis techniques (such as formal hypnosis or hypnotic conversation, practiced by qualified professionals). Percutaneous cryoablation was performed under strict aseptic conditions in a dedicated IR suite, by a team of 5 experienced interventional radiologists, all of whom were trained for 5–15 years in percutaneous tumor ablation and in various ablation techniques including cryoablation. The procedure was performed under US or computed tomography (CT) depending of the location and the size of the lesion. One or more 13Gauge cryoprobes (V probe, Varian, USA, or ProSense, IceCure Medical, Israel) were used for each patient. After subcutaneous injection of lidocaine and skin incision, the probes were positioned preferentially in the center of the target along its long axis, with a pubic entry point to minimize the visibility of the entry point scar. If several probes were used for a same nodule, they were spaced of 15–20mm. If several nodules were present, they were treated simultaneously with different probes. The replacement of the probes and overlapping ablation was obtained as needed based on radiologist’s evaluation to cover the whole lesion. A protocol of two cycles for up to 10 minutes of freezing each spaced by 10 minutes of passive thawing was initially planned, but each cycle duration and power settings were adapted to the size of the nodule targeted with the endpoint to cover the totality of the nodule while preserving the skin and the digestive tract. No active thawing was used at any time. The cryoprobes were removed after passive thawing. If needed, protective techniques were used such as skin warming, hydrodissection, and carbodissection. A peri-procedural intravenous steroid injection (dexamethasone 8mg) was performed, and oral non-steroidal anti-inflammatory drugs (ibuprofen) were prescribed for 4 days after the procedures. Patients were informed of normal transient local inflammatory reaction that may be observed after cryoablation and were given an emergency number in case needed. Pain-free survival, defined as time from cryoablation to recurrence of pain or to the last day of follow-up, was the primary endpoint. Deep pelvic endometriosis pain was distinguished from parietal pain as the latter is perceived as more intense and superficial, with a palpable trigger point and a nodule. In patients with associated deep endometriosis pain, only parietal pain related to the nodule or related to the procedure was evaluated. Pain was systematically assessed by the visual analogue scale (VAS) during the pre- and post- procedure consultation planned at 1, 3, 6, 12, 24 and 36 months. Relapse was arbitrarily defined as a re-escalation of pain on the VAS during follow-up. As secondary endpoint, adverse events were evaluated in consultation with the SIR Adverse Event Classification System [ 23 ]. Aesthetic sequelae and entry-point scar were recorded during follow-up consultations. An imaging follow-up was carried out with contrast enhanced MR imaging between 1 and 3 months (T2-weighted and T1-weighted sequences were obtained before and after administration of an intravenous bolus of 0.1 mmol/kg (0.5 mmol/mL) of gadoteric acid (Dotarem; Guerbet, Roissy, France) at a rate of 2 mL/s and followed by a 20-mL saline flush). Patient Global Impression of Change (PGIC) scale was used as a marker of quality of life and recorded at the last follow-up [ 24 ]. For analysis of pain-free survival, a Wilcoxon matched-pairs signed rank test was performed and P < 0.05 was considered statistically significant. The survival curves were constructed using the Kaplan-Meier method. To identify the recurrence rate in patients with associated deep endometriosis, Fisher’s exact test was used. The primary efficacy rate is defined as the percentage of pain relief estimated by the patients following the initial procedure. The secondary efficacy rate is defined as successful repeat ablation following recurrence of pain. The efficacy rate of cryoablation to avoid secondary surgery was also evaluated. Data was entered into a worksheet for storage and extraction/analysis (Data analysis: Prism 9; GraphPad Software, San Diego, USA). All images were retrospectively reviewed by a radiologist with 8 years’ experience.

Technical characteristics are summarized in Table 2 . A single session of cryoablation was carried out under hypnosis, sedation and local anesthesia for 18 patients (42.9%, 18/42) and under general anesthesia for 24 patients (57.1%, 24/42). Twenty-eight patients (66.7%, 28/42) were managed as outpatients while 14 patients were hospitalized overnight per patient request (median length of stay for these 14 patients: 1 day [ 1–1 ]). Mostly US image guidance was used (83.3%, 35/42) with US alone in 61.9% (26/42) and in combination with CT in 45.2% (19/42). A median of one cryoablation probe per nodule was used per cryoablation (range 1–3). The hydrodissection technique was systematically performed (100%, 47/47), while carbodissection was performed in 13 (27.6%, 13/47) lesions. When carbodissection was performed, CT was systematically used. Technical success was 100% (47/47). For procedures that used CT as a guidance modality (44.6%, 21/47), the median total dose length product (DLP) was 1447.5 mGy-cm (SD: 1175). Of the 42 patients treated, 2 patients (4.8%, 2/42) were lost to follow-up: one patient before the first M1 follow-up appointment, and a second patient before the second M3 follow-up appointment (the pain had decreased from 8 to 6 out of 10 at M1 appointment). Those two patients were unsuccessfully contacted at regular intervals until the data collection. Median follow-up was 13.5 months [1.1 – 37.7]. Three patients (7.1%, 3/42) had surgical resection of the nodule after cryoablation: one had a disappearance of pain (VAS 0/10) but the nodule was still palpable and the patient requested surgical resection, another one had a disappearance of pain (VAS 0/10) but a catamenial umbilical external bleeding, and the last patient had an insufficient decrease of pain at the first month consultation (VAS going from 8/10 before cryoablation to 6/10) and requested surgical resection. The entry-point scar was no longer visible in all patients. Efficacy rate of cryoablation to avoid secondary surgery is therefore 92.8% (39/42) per patient and 93.6% (44/47) per nodule treated. Median VAS before procedure was 8/10 [IQR: 7–9] and dropped to 0/10 [0–1] at the last follow-up (Wilcoxon test, P < 0.0001). The evolution of VAS during the systematic follow up is shown on Figure 1 . Median PGIC score recorded at last follow-up was 1/7 [ 1 – 2 ]. Four patients (9.5%, 4/42) had recurrence of their pain after the procedure: 3 patients underwent a new cryoablation session after MRI control showed signs suggestive of recurrence (hemorrhagic spots, intense enhancement after injection) and no longer had pain afterwards with PGIC scores at last follow-up ranging from 1 to 2. One patient had a recurrence of her pain one year after cryoablation (VAS at 9/10 before intervention dropping to 0/10 in the following year then returning to 3/10 at M12 last follow-up). The primary efficacy rate of cryoablation is therefore 90.4% (38/42) and secondary efficacy rate 97.6% (41/42). According to the Kaplan-Meier method, the median pain-free survival rates are 93.75% [95% CI, 77.25–98.4] at 6 month, and 82.72% [58.8–93.45] at 12 month, 24 month and 36 month, respectively ( Figure 2 ). Despite the low recurrence rate, there does not appear to be a correlation between recurrence of the pain and the presence of associated proven deep endometriosis (Fisher’s exact test P > 0.9999). Four patients (9.5%, 4/42) had adverse event in the days following the procedure, with one severe (1/42). Two patients had mild adverse events (Grade 1): a skin anesthesia of about 2cm 2 next to the ablation area. One patient had a 2 nd -degree skin burn that resolved after one week of local care (Grade 2). This was due to skin involvement of the endometriosis. One patient had a severe adverse event (grade 3) on day 15 (2 nd -degree skin burn and peritonitis from a probable small bowel injury) which required local skin care and five days hospitalization with intra-veinous antibiotic therapy. This patient presented an extensive parietal and cutaneous involvement, inoperable, with digestive adhesions, making the cryoablation of high risk of adverse events ( Figure 3 ). During follow-up, this patient had no more pain (VAS 0/10 on last follow-up versus 8/10 before procedure), with a PGIC score of 1/7.

Percutaneous cryoablation of EE appears safe and effective. Cryoablation presents an excellent safety profile with only one severe adverse event reported in this retrospective study performed over 5 years. One of the main advantages of cryoablation is to avoid the coagulative necrosis usually observed with heat-based techniques such as radiofrequency ablation or high intensity focused ultrasound (HIFU), which enables collagen structures to be spared and tissue granulation avoided, leading to improved esthetic outcomes [ 18 , 25 ]. Another advantage of cryoablation over other local heat-based therapies is that while delivering tissue-ablative freezing temperatures [ 26 ], the ice is visible under real-time US and CT imaging guidance [ 27 , 28 ]. The risks of injuring the skin or bowel is therefore low, and the use of other adjunctive techniques, such as tissue dissection, may further limit these risks [ 29 – 31 ]. Moreover, as it was already demonstrated but confirmed in this study, cryoablation may be performed safely under local anesthesia and on an outpatient basis, with a low number of cryoprobes, reducing the length of hospitalization and the postoperative recovery [ 27 , 32 – 36 ]. However, although cryoablation has been used in treatment of dysfunctional uterine bleeding to achieve destruction of the entire functional endometrium, additional data is needed to better understand and design future comparative clinical trials [ 37 ]. Among the patients in this series who had surgery after cryoablation, one of the patients had concerns with the size of the nodule remaining after treatment. It is therefore critical to evaluate the underlying role of vascular injuries, inflammation, or immune response occurring after cryoablation of EE and their evolution over time to better understand the postoperative symptoms [ 38 ]. In this study, primary and secondary efficacy were high, similar to surgery alone, for the local control of extraperitoneal endometriosis [ 3 , 39 ]. As an alternative to surgery or in case of inoperable soft tissue tumors, cryoablation has emerged as a second-line minimally invasive therapeutic option in many indications such as desmoid tumors or vascular malformations [ 19 , 33 ]. Percutaneous cryoablation of EE was effective on pain with satisfactory quality of life evaluation after treatment. However, patients must be informed and aware of the potential outcomes after cryoablation as pain and discomfort may be multifactorial in case of concomitant deep endometriosis or after surgery [ 40 ]. This fact is demonstrated in three of the patients who underwent further surgery despite having achieved pain relief. For lesions associated with deep endometriosis or in the case of large and infiltrative lesions, cryoablation should not be proposed alone but eventually in combination with hormonal or surgical treatments, which needs to be further evaluated. A very close cooperation with gynecologists is therefore needed to manage the patient holistically and ensure adequate follow-up. Pain management may also require multidisciplinary evaluation to control symptoms such as postsurgical neuropathic pain may be the cause of symptoms instead of the EE itself. Lesions located in the umbilicus are more challenging to treat than any other as skin and bowel loops are often very close to the targeted area. In this location, MR guidance may help improve outcomes [ 41 ]. However, this study confirmed that cryoablation may have a role for patients presenting multiple nodules which can be treated efficiently in one single procedure, as carried out for 2 patients, avoiding multiple interventions or extensive surgery, as was suggested for desmoid tumors [ 34 ]. The retrospective design is a limit of this study performed in a single institution. A case-by-case discussion was performed in specific multidisciplinary team meetings, which may have introduced a selection bias. A biopsy was not performed systematically to rule out other diagnoses such as sarcoma or desmoid tumors as some women had a history of endometriosis or clinical and radiological features of EE as suggested by Hensen et al. [ 20 ]. To conclude, percutaneous cryoablation demonstrated its safety and efficacy for the treatment of extraperitoneal endometriosis. Cryoablation may be proposed as an alternative to surgery to limit recovery time in active or child-bearing aged women with similar outcomes. However, cryoablation must be evaluated in multidisciplinary committees to ensure adequate selection and follow-up of women. Prospective clinical trials may now be designed to further evaluate the technique and incorporate it in future guidelines of therapeutic management of extraperitoneal endometriosis.

Endometriosis affects 190 million women of reproductive age worldwide [ 1 ]. This chronic disease is associated with severe, life-impacting symptoms, but many individuals suffering from it have limited awareness of the condition causing a lengthy delay between onset of symptoms and diagnosis [ 2 ]. Extraperitoneal endometriosis (EE) may be observed incidentally but more often on surgical scars in women with or without history of endometriosis [ 3 ]. Clinically, the cyclic fluctuation of symptoms of EE can be observed in around half of patients, but ultrasound or MR imaging may help to further confirm the diagnosis [ 4 , 5 ]. To avoid the progression of EE while reducing symptoms, therapeutic options have been thus far limited to hormonal agents or wide surgical excision [ 6 – 8 ]. During the last decade, various minimally-invasive therapies have been evaluated successfully to provide local control of EE [ 9 – 16 ]. Among these, cryoablation has been suggested to be a promising option to treat abdominal wall endometriosis with satisfactory outcomes and low morbidity [ 15 – 18 ]. Cryoablation of soft tissue lesions may be performed safely as the ice produced remains visible under the imaging techniques used in interventional radiology to guide the procedure. This enables precise assessment of the margins of ablation and thus helps to reduce the risks of adverse events [ 19 ]. The purpose of this retrospective study performed in a single institution was to evaluate the pain-free survival and the outcomes of percutaneous image-guided cryoablation of symptomatic extraperitoneal endometriosis.