Long-term outcome of transendoscopic oesophageal ...

3 downloads 0 Views 562KB Size Report
Oct 1, 2017 - transendoscopic oesophageal mass ablation in dogs with spirocercosis- ..... irregular, with areas of haemorrhage and necrosis, while benign.
Downloaded from http://veterinaryrecord.bmj.com/ on May 1, 2017 - Published by group.bmj.com

Paper

Paper Long-term outcome of transendoscopic oesophageal mass ablation in dogs with Spirocerca lupi-associated oesophageal sarcoma A. Shipov, G. Kelmer, E. Lavy, J. Milgram, I. Aroch, G. Segev Oesophageal sarcoma is a potential sequel of Spirocerca lupi infection. Oesophageal mass excision can be performed by open chest surgery. The objectives of this observational study were to evaluate the feasibility, short-term morbidity and long-term outcome of transendoscopic oesophageal mass ablation in dogs with spirocercosis-associated oesophageal neoplasia. A 9 mm video-endoscope and laser or electrocauterisation were used to debulk the oesophageal mass. Long-term follow-up was done by telephonic interviews. Fifteen dogs were included. The median tumour size was 5 cm (range 3.5–9). The median procedure time was 75 minutes (range 35–165) and was deemed successful in 12/15 dogs (80 per cent). Recovery was uneventful in all dogs. Immediate complications included oesophageal damage (two dogs) oesophageal perforation (one dog) and a focal thermal damage (one dog). The median hospitalisation time of all dogs was less than one day, with all but two discharged on the procedure day. The median survival time, available in nine dogs that were followed, was 202 days (range 51–691). Four of these dogs (44 per cent) survived more than six months, of which three survived more than one year. In conclusion, transendoscopic oesophageal mass ablation might be considered an alternative, palliative procedure for open-chest oesophageal surgery. It has comparable long-term survival, lower morbidity, short hospitalisation time and relatively low cost.

Spirocerca lupi is a nematode-infecting canidae, and is most prevalent in tropical and subtropical areas (Mazaki-Tovi and others 2002, van der Merwe and others 2008). S. lupi eggs are shed in the host’s faeces and are ingested by coprophagous beetles, in which nematode larvae (L1) develop into infective larvae (L3). When infected beetles are ingested by a dog, or infected paratenic hosts are preyed by dogs, larvae are released into the gastric lumen, where they penetrate the gastric mucosa and invade the gastric arterial walls and migrate to the aortic wall. Larvae then migrate through the thoracic cavity, finally reaching the distal oesophageal wall, where they mature and sexually reproduce. Their presence induces formation of fibrous tissue nodules. L1-containing eggs are then shed through an orifice in the nodule into the oesophageal lumen and are then passed in the faeces (van der Merwe and others 2008). Oesophageal sarcoma occurs in 8–26 per cent of the cases (Brodey and others 1977, Dvir and others 2001, Ranen and others 2004a, 2008, van der Merwe and others 2008). These tumours grow locally, induce inflammation (Nivy and others Veterinary Record (2015) A. Shipov, DVM, Diplomate ECVS, G. Kelmer, DVM, Diplomate ECVS, E. Lavy, DVM, Diplomate ECVPT, J. Milgram, DVM, Diplomate ECVS, I. Aroch, DVM, Diplomate ECVIM-CA, G. Segev, DVM, Diplomate ECVIM-CA, Koret School of Veterinary Medicine,

doi: 10.1136/vr.103356 The Hebrew University of Jerusalem, Rehovot, Israel E-mail for correspondence: [email protected] Provenance: not commissioned; externally peer reviewed Accepted September 7, 2015

2014) and often metastasise to the lungs (Ranen and others 2008). The molecular events and the time-course involved in transformation of the fibrous nodules to neoplasia are unknown (Dvir and others 2009, Dvir and Clift 2010). The clinical signs of oesophageal spirocercosis, benign or malignant, often include vomiting, regurgitation, weight loss, hypersalivation and sialoadenomegaly (Mazaki-Tovi and others 2002, van der Merwe and others 2012). Progressive growth of spirocercosis-associated oesophageal tumours often leads to more severe clinical signs, including melena, haematemesis, resulting in pale mucous membranes and anaemia, as well as respiratory signs due to aspiration pneumonia and pulmonary metastasis (Ranen and others 2008). Benign spirocercosis is completely responsive to macrolidic lactone (avermectins and milbemycin-oxime) treatment (Lavy and others 2002, Kok and others 2011). However, with malignant transformation of the oesophageal lesions, antihelminthic therapy is ineffective, and treatment options are only palliative (Ranen and others 2004b). Oesophageal resection and anastomosis, or partial oesophagectomy, with or without chemotherapy, are the most commonly reported surgical treatment options for such neoplastic oesophageal masses; however, these procedures are associated with high morbidity and mortality rate (Ranen and others 2004b, 2008). In six dogs with spirocercosis-associated oesophageal neoplasia, resection of the lesion by partial oesophagectomy with 1–2 cm margins was performed as a palliative measure and was associated with a relatively long hospitalisation (mean 6.3 days, range 3–11); however, in 5/6 dogs, postoperative life quality was satisfactory, and survival time was 2–18 months in five dogs and >20 months in October 10, 2015 | Veterinary Record

Downloaded from http://veterinaryrecord.bmj.com/ on May 1, 2017 - Published by group.bmj.com

Paper one. Adjunctive doxorubicin-based chemotherapy was administered to 5/6 dogs in that study (Ranen and others 2004b). In a more recent study, of 32 dogs with S. lupi-associated oesophageal neoplasia, 9/19 dogs (47 per cent) undergoing oesophageal surgery did not survive to discharge, three of which were euthanased during or shortly after surgery, as the tumours were deemed inoperable, and six died during the perioperative period (Ranen and others 2008). The median survival time of the 10 surviving dogs, 6 of which also received adjunctive chemotherapy, was 278 days (range 60–610) (Ranen and others 2008). Transendoscopic oesophageal mass ablation using laser and electrocautery was described in two dogs with spirocercosisassociated oesophageal neoplasia (Yas and others 2013). Due to the limited number of cases, robust conclusions regarding the procedure’s complication rate and long-term outcome, especially in comparison to open-chest oesophageal surgery, cannot be made. Therefore, the aims of the present study were to evaluate the feasibility of transendoscopic oesophageal mass ablation using laser or polypectomy snare electrocautery in a larger cohort of dogs with spirocercosis-associated oesophageal neoplasia and to assess the short-term morbidity and long-term outcome.

Materials and methods Selection of dogs Dogs diagnosed during years 2012–2014 at the Hebrew University Teaching Hospital with oesophageal neoplasia, based on gross endoscopic examination, were prospectively and consecutively selected. Transendoscopic mass ablation was performed as a palliative procedure only if the owners declined open-chest oesophageal surgery and signed a consent form. Consequently, transendoscopic oesophageal mass ablation was attempted in all dogs included, regardless of tumour size or nature of its attachment to the oesophageal wall (i.e. diffusely infiltrative or connected to the oesophageal wall via a stalk).

Procedure In all cases, a 9 mm video-endoscope (Fujinon, Fujifilm Global, Japan) was used. Tumour size was evaluated by measuring the distance between the most proximal and distal parts of the tumour using the markings placed on the endoscope. To decrease the risk of oesophageal perforation during the procedure, 5–15 ml of sterile saline (first three cases) or hetastarch (Fresenius-Kabi, Germany; in all subsequent cases) were injected submucosally at the base of the tumour, to isolate it from the normal oesophageal tissue, using an injection needle introduced through the endoscope’s working channel. Laser ablation was performed in the first dog using a neodymium-yttriumaluminium garnet laser (Nd:YAG 100 W, Sharplan 2000, Lumenis, Israel), and using a diode laser (Diode 60 W, Vetrix, UK) later on. The laser fibre was introduced through the endoscope’s biopsy channel, and used in a continuous mode, with the power initially set at 20 W, and increased up to 60 W if needed. The laser beam was discharged at short, three- to five-second bursts using a non-contact mode, assuring a distance of approximately 5–10 mm from the tissue, until the mass was excised (Fig 1). Electrocautery ablation was performed using a polypectomy snare with a 6 cm diameter loop (MTW, Germany) connected to a cutting-cautery unit (Valleylab, SSE2L, Colorado, USA) using a monopolar mode. The snare was introduced and manipulated to encircle the mass, avoiding normal oesophageal mucosal entrapment. Intermitted current was then delivered, until the mass was resected. In the last three dogs, to improve visualisation of the mass-oesophageal wall attachment, the mass was retracted using a wire loop. The loop was introduced percutaneously into the gastric lumen under endoscopic guidance, pulled into the oesophagus using biopsy forceps and manipulated to encircle the entire oesophageal mass. The mass then was pulled caudally by applying gentle traction to the wire loop until the mass-oesophageal wall attachment was completely visible. Veterinary Record | October 10, 2015

The procedure was considered successful when ≥90 per cent of the visible oesophageal mass was resected.

Follow-up Follow-up endoscopy was performed in four dogs. Long-term follow-up was performed by telephone interviews with the owners. Each owner was asked to describe any postprocedure complication and its nature, and to grade their satisfaction with the procedure on a scale of 1–10. Owners’ stratification was based on subjective assessment of the dog’s quality of life after the procedure. Once gastrointestinal clinical signs recurred, owners had to make the decision whether to perform additional attempt to respect the mass or to euthanase the dog. When clinical signs were associated with lung metastasis, dogs were euthanased. Owners were also asked whether they would elect to perform this procedure in the future, given the same circumstances.

Statistical methods Descriptive statistics were used in continuous and categorical parameters. The Kaplan-Meier survival analysis was used to assess the long-term outcome. Dogs lost to follow up were censored, and analysis was based on the last information obtained from the owners. Analyses were performed using a statistical software package (SPSS V.17.0 for Windows, SPSS, Illinois, USA).

Results The study included 15 dogs (six males and nine females) endoscopically diagnosed with spirocercosis-associated neoplasia, including mixed-breed (seven dogs), labrador or golden retriever (four) and other breeds (four). The median age was 8.7 years (range 3.3–14). The median tumour size, measured during endoscopy, was 5 cm (range 3.5–9). None of the dogs had radiographic evidence of pulmonary metastasis at the time of diagnosis. The procedure was deemed successful in 12/15 of the cases (80 per cent). In one dog, the oesophageal mass was large (8 cm) and could not be adequately differentiated from the normal oesophageal mucosa, and therefore, the attempt to resect it was discontinued shortly after initiating the procedure. In two dogs, with tumour diameters of 7 and 9 cm, attempts were made to resect the masses using the laser, lasting 120 and 150 minutes, respectively, but were unsuccessful. Transendoscopic mass ablation was performed using the laser in five dogs, by snare electrocautery in six dogs and by their combination in three dogs. The median procedure time in the cases in which mass debulking was performed was 75 minutes (range 35–165). Recovery was uneventful in all dogs. Immediate complications included oesophageal damage in three dogs and a focal thermal damage to the skin on the lateral aspect of the right elbow in one dog. In that dog, electrocautery was used and the thermal damage occurred in the contact area between the skin and the ground return pad. The final size of this skin lesion was 4 cm. It was treated topically with silver-sulfadiazine cream (silverol; Teva, Israel) and bandaging, and progressively healed, over two weeks. One of the three dogs with oesophageal damage had a distal oesophageal perforation and was consequently euthanased. In a second dog, a moderate, self-limiting oesophageal haemorrhage occurred, which did not require blood transfusion. This dog recovered uneventfully, and postprocedure follow-up haematocrits were stable. The dog was discharged on that same day. In the third dog, several centimetres of normal oesophageal mucosa were removed along with the mass, when the latter was pulled out using the polypectopy snare. A percutaneous endoscopically placed gastrostomy (PEG) tube was placed, and the dog was hospitalised for four days with supportive symptomatic therapy. Two weeks later, follow-up oesophagoscopy showed complete oesophageal mucosal healing, with no evidence of recurrence of the mass. This dog was still alive at 691 days post procedure, when this report was being written. In the remaining dogs, local haemorrhage and oesophageal mucosal damage were minimal.

Downloaded from http://veterinaryrecord.bmj.com/ on May 1, 2017 - Published by group.bmj.com

Paper (a)

(b)

FIG 1: (a) Endoscopic view of the distal oesophagus after transoesophageal laser debulking of a Spirocerca lupi-associated tumour; (b) the resected mass

The median hospitalisation time of all dogs was less than one day. All but two dogs were discharged on the procedure day, while in these two the hospitalisation periods were one and four days. Oesophageal mass histopathology was available in all 12 dogs in which the oesophageal masses were resected, 10 of which had osteosarcoma and 2 had fibrosarcoma. Follow-up endoscopy was performed in four dogs. In one, endoscopy was performed 30 days post procedure due to recurrence of clinical signs and revealed a 2 cm oesophageal mass, which was resected using cauterisation. However, clinical signs recurred, and this dog was euthanased 51 days post the first procedure. In another dog, with oesophageal damage, described above, a 14-day follow-up endoscopy showed no evidence of recurrence. In the remaining two dogs, follow-up endoscopies were performed at 60 and 90 days post procedure and showed no evidence of recurrence. Long-term follow-up was available in 9/11 dogs that underwent the procedure and survived to discharge. At the follow-up time, three dogs were still alive and six had been euthanased, five of which due to recurrent gastrointestinal signs and one due to pulmonary metastasis. No procedure-associated complications were noted by any dog owner. The median survival time of these nine dogs was 202 days (mean 259; range 51–691; Fig 2). Four dogs (44 per cent) survived more than six months, of which three survived more than one year. The median owner satisfaction score was 8 (range 3–10), and 8/9 stated that given the same

1.0 Censored

Survival rate

0.8

0.6

0.4

0.2

0.0 0

200

400

600

Days from discharge

FIG 2: Kaplan-Meier postdischarge survival curve of dogs that underwent laser or electrocauterisation transoesophageal resection of spirocercosis-associated oesophageal sarcoma. Dogs were censored if were lost to follow-up. Censored dogs are depicted by small horizontal lines

circumstances they would have elected to perform the procedure.

Discussion This study demonstrates that oesophageal tumour debulking using transendoscopic approach is feasible, associated with relatively low complication rate and high perioperative survival, and a median survival time comparable to that of dogs undergoing oesophageal tumour excision via open thoracic surgery. Oesophageal neoplasia is rare in areas not endemic to S. lupi (Kyles 2012). Malignant transformation of oesophageal lesions to sarcoma occurs in 8–26 per cent of dogs with spirocercosis (Brodey and others 1977, Dvir and others 2001, Ranen and others 2004a, van der Merwe and others 2008). Therefore, in endemic areas for spirocercosis, caudal oesophageal sarcomas are assumed to be associated with S. lupi infection, although this association is presumptive, as in such sarcomas adult worms are most often absent (van der Merwe and others 2008). Antemortem diagnosis of S. lupi-associated oesophageal neoplasia is often challenging. Several studies have investigated several non-invasive diagnostic modalities to differentiate antemortem between benign nodules and oesophageal neoplasia (Nivy and others 2014); however, a definitive diagnosis can only be made by histopathology of representative oesophageal mass tissue sections. Endoscopically obtained biopsies are limited because mostly only superficial samples may be obtained, and these are therefore more often inconclusive due to the inflammation and necrosis associated with the neoplasia. Thus, a tentative diagnosis often has to be made based on the gross endoscopic appearance of the mass. Neoplastic masses are relatively large and irregular, with areas of haemorrhage and necrosis, while benign masses are smaller, smooth and rarely bleed, and often contain a ‘nipple-like’ protuberance, through which the female nematode sheds its eggs (van der Merwe and others 2008). Diagnosis of oesophageal neoplasia based on gross endoscopic appearance of the mass is not definitive, but is highly sensitive. Although the accuracy of gross endoscopic diagnosis of spirocercosis-associated oesophageal neoplasia has not been studied thoroughly, in a previous study, there was 100 per cent agreement between the gross endoscopic findings of large superficially necrotic, ulcerative and haemorrhaging oesophageal masses and oesophageal malignancy, based on representative tissue samples obtained for histopathology (Dvir and others 2001). The present study was not designed to assess the diagnostic performance of gross endoscopic appearance in spirocercosis-associated oesophageal lesions. Nevertheless, findings show that it was accurate in 100 per cent of cases compared with the histological diagnosis of the resected tumours. Therefore, it seems that in cases of oesophageal tumours in dogs the gross endoscopic appearance is highly suggestive of neoplasia. Until recently, oesophageal resection and anastomosis, or partial oesophagectomy, with or without adjunctive chemotherapy, were the only available palliative treatment options for October 10, 2015 | Veterinary Record

Downloaded from http://veterinaryrecord.bmj.com/ on May 1, 2017 - Published by group.bmj.com

Paper S. lupi-induced oesophageal sarcoma (Ranen and others 2004b, 2008). These palliative procedures were shown to be feasible and provided good life quality in most cases, and in dogs surviving to discharge the median survival time was 9.1 months (range 2–20 months) (Ranen and others 2004b, 2008). However, these procedures are invasive and are associated with pain, high mortality, prolonged hospitalisation and high cost (Ranen and others 2004b, 2008). Therefore, and since this treatment is mostly only palliative, owners often elect to euthanase their dog once oesophageal neoplasia is diagnosed, rather than pursue surgery. In a previous report of two dogs with spirocercosis-associated oesophageal neoplasia, transendoscopic oesophageal mass ablation using either laser or snare electrocautery had no complications, and hospitalisation time was short (1–2 days), and therefore, treatment cost was lower compared with open thoracic oesophageal surgery (Ranen and others 2004b, Yas and others 2013). The present study demonstrates that transendoscopic oesophageal mass ablation is feasible in most (80 per cent) cases and is associated with an acceptable complication rate. In 3/15 cases, the oesophageal mass could not be removed using the technique since these masses were relatively large and their communication with the oesophageal mucosa was diffuse, posing a high, unacceptable risk of oesophageal perforation. It is plausible that with gaining further experience with this technique the success rate will increase. Submucosal saline, hetastarch or dye material injection under the oesophageal mucosa will facilitate resection of large masses, while decreasing the risk of oesophageal damage. Since spirocercosis-associated masses mostly occur in the distal oesophagus (Dvir and others 2001), there is constant movement of the lesion during the procedure due to cardiac and diaphragmatic activity, varying depending on the location of the mass relative to the heart and the heart rate. This movement decreases the laser cutting accuracy, thereby increasing the risk of oesophageal wall damage and prolonging the procedure. Despite the extensive vascularisation within the tumours, haemorrhage was uncommon and was deemed severe in only one dog. Potentially, combining laser and electrocauterisation in the procedure will substantially decrease the risk of haemorrhage. One immediate complication, documented in three dogs herein, was oesophageal wall damage. In one dog in which the oesophagus was perforated, the owners refused further treatment and elected euthanasia. In a second dog, PEG tube placement with supportive care in hospitalisation allowed oesophageal healing, which was complete as noted up on a two-week follow-up endoscopy. Therefore, if the procedure results in oesophageal wall damage, these measures should be considered a good therapeutic option. Dog owners should be made aware, prior to the procedure, of this potential complication, as well as the feasible option to treat oesophageal wall damage with PEG tube placement. The risk of oesophageal perforation and owners’ declining feeding tube placement or corrective surgery might dictate a more conservative mass debulking, which might negatively affect the outcome. Another complication, observed in one dog, was thermal dermal injury to the right elbow. The time in which electrocauterisation was used in this procedures was relatively long, compared with that used during routine surgery, likely increasing the risk of electrical burn. The authors suspect that this complication occurred due to a poor contact between the ground return pad and the dog, resulting in high current density at the contact site. Following this event, only disposable ground plates were used instead of a reusable metal pad, the hair in the contact area was thoroughly clipped and gel was applied, ensuring complete pad contact with a large skin surface area. With implementation of these measures, no additional thermal damage cases have occurred. The number of dogs included in this study and the variable tumour size precludes a comprehensive comparison between electrocautery and laser mass resection, and occasionally, these Veterinary Record | October 10, 2015

two measures were combined. Spirocercosis-associated oesophageal sarcomas might be mineralised (Dvir and others 2001), and when mineralisation is diffuse, electrocauterisation is probably ineffective or less effective. The authors’ subjective assessment is that when oesophageal masses are relatively large (>4 cm) or diffusely mineralised, the use of laser is more effective in resecting the mass, while using polypectomy snare electrocauterisation should be reserved for smaller masses, connected to the oesophagus wall by a relatively thin stalk. Regardless of the method used, once resected, the mass can be easily pulled out using the snare. The long-term survival time distribution in this study was somewhat bimodal, with some dogs surviving for relatively short periods (less than three months), while others were free of clinical signs for more than one year. This is in agreement with previous findings in surgically treated dogs with oesophageal sarcoma, with 3/6 dogs surviving two to seven months, and three for more than one year (Ranen and others 2004b). Because all dog owners in this study declined open-chest oesophageal surgery, but elected the transendoscopic mass resection as a last measure prior to euthanasia, the procedure was either attempted or successfully performed even if the masses were large or were diffusely attached to the oesophageal wall. Likely, selecting only cases with relatively small oesophageal masses, with limited communication with the oesophageal mucosa (i.e.