A novel ultrasound-based algorithm for detection of pancreatic stents placed for prophylaxis of post-ERCP pancreatitis: a prospective trial.

PURPOSE: Before removal of retained pancreatic stents placed during endoscopic retrograde cholangiopancreatography to avoid post-ERCP pancreatitis an imaging is recommended. The aim of the present study was to evaluate a new ultrasound-based algorithm. MATERIALS AND METHODS: Patients who received a pancreatic stent for PEP prophylaxis were included. Straight 5Fr (0.035inch) 6cm stents with an external flap that were visualized by ultrasound were removed endoscopically with no further imaging. If the ultrasound result reported the stent to be dislodged or was inconclusive, X-ray imaging was performed. The endpoints were positive and negative predictive value, specificity, sensitivity, and contingency coefficient between ultrasound and X-ray and/or endoscopy. RESULTS: In the present study, 88 patients were enrolled. X-ray was performed in 23 (26%) patients. Accordingly, the ultrasound algorithm saved an X-ray examination in 65 cases, leading to a reduction of 74%. Stents were retained in 67 patients (76%) and visualized correctly by ultrasound in 54 patients with a sensitivity of 81%. The positive predictive value was 83%. Specificity was 48% because ultrasound described 10/21 dislodged stents correctly. The negative predictive value was 43% as 10/23 stents were correctly classified as dislodged by ultrasound. In 11 patients (13%), esophagogastroduodenoscopy was performed even though the pancreatic stent was already dislodged. CONCLUSION: A novel ultrasound-based algorithm reduced the need for X-ray imaging by three quarters. To avoid unnecessary endoscopic examinations, the algorithm should be implemented with a learning phase and procedures should be performed by experienced examiners. An important limitation might be the stent lengths, as shorter stents might be more difficult to visualize by ultrasound. __________________ Hintergrund: Aktuell wird vor der Entfernung von prophylaktisch gelegten Pankreasstents nach einer endoskopischen retrograden Cholangiopankreatikographie eine Bildgebung empfohlen. Ziel der vorliegenden Studie war es, einen neuen ultraschallbasierten Algorithmus zu evaluieren. MATERIAL UND METHODEN: Eingeschlossen wurden Patienten nach prophylatischer Pankreasstentanalage . Gerade 5 Fr-Stents (0.035 inch) mit 6 cm Länge vom externen Flange, die mittels Ultraschall sichtbar waren, wurden endoskopisch ohne weitere Bildgebung entfernt. Wenn das Ultraschallergebnis den Stent als disloziert beschrieb, wurde eine Röntgenaufnahme durchgeführt. Die Endpunkte waren der positive und negative Vorhersagewert, die Spezifität, Sensitivität und der Kontingenzkoeffizient zwischen Ultraschall und Röntgen und/oder Endoskopie. Ergebnisse: 88 Patienten wurden in die Studie eingeschlossen. Bei 23 (26%) Patienten musste eine Röntgenaufnahme durchgeführt werden. Entsprechend hat der Ultraschallalgorithmus in 65 Fällen (74%) eine Röntgenuntersuchung eingespart. Stents waren bei 67 Patienten (76%) verblieben und wurden bei 54 Patienten korrekt mit einer Sensitivität von 81% mittels Ultraschall visualisiert. Der positive Vorhersagewert betrug 83%. Die Spezifität betrug 48%, da der Ultraschall 10/21 dislozierte Stents korrekt beschrieb. Der negative Vorhersagewert betrug 43%, da 10/23 Stents korrekt als disloziert klassifiziert wurden. Bei 11 Patienten (13%) wurde eine Ösophagogastroduodenoskopie durchgeführt, obwohl der Pankreasstent bereits disloziert war. Fazit: Ein ultraschallbasierter Algorithmus reduzierte den Bedarf an Röntgenbildgebung um drei Viertel. Um unnötige endoskopische Untersuchungen zu vermeiden, sollte der Algorithmus mit einer Lernphase implementiert und das Verfahren von erfahrenen Untersuchern durchgeführt werden. Eine wichtige Einschränkung könnte die Länge der Stents sein, da kürzere Stents mit Ultraschall schwieriger zu visualisieren sein könnten.

Radiofrequency ablation versus hybrid argon plasma coagulation in Barrett's esophagus: a prospective randomised trial.

INTRODUCTION: Radiofrequency ablation (RFA) and hybrid argon plasma coagulation (H-APC) are established thermal ablation techniques for eradicating Barrett's esophagus after endoscopic resection. This study aimed to compare RFA with H-APC in relation to safety, effectiveness and eradication rates. METHODS: After endoscopic resection, patients were randomly assigned to H-APC or RFA. A simplified H-APC technique was applied at 60 W. RFA was used with a 90° focal catheter and a simplified protocol of 12 J/cm2 × 3 or with a Halo 360° balloon and 10 J/cm2/cleaning/10 J/cm2. Eradication rates and adverse events were recorded. Patients received follow-up examinations after 3, 6, 12 and 24 months. RESULTS: One hundred and one patients were finally included in the study (RFA N = 47, H-APC N = 54). The median follow-up period for short-term was 6.0 (CI 5.4-6.9) months and for long term 21 (CI 19.2.5-22.7) months. In total 211 ablations were performed. The eradication rates after long-term follow-up were 74.2% in the RFA group and 82.9% in the H-APC group. Post-interventional pain was significantly greater in the RFA group, with a mean score of 4.56/10 and duration of 7.54 days, in comparison with a mean score of 2.07/10 over 3.59 days in the H-APC group. Stenoses requiring intervention were noted in 3.7% of patients in the H-APC arm and 14.9% of those in the RFA arm. CONCLUSIONS: Both ablation techniques have good results in relation to the eradication rate, with a slightly better outcome in the H-APC group. The severity and duration of pain were significantly greater in the RFA group.

Artificial intelligence-assisted staging in Barrett's carcinoma.

BACKGROUND: Artificial intelligence (AI) is increasingly being used to detect neoplasia and interpret endoscopic images. The T stage of Barrett's carcinoma is a major criterion for subsequent treatment decisions. Although endoscopic ultrasound is still the standard for preoperative staging, its value is debatable. Novel tools are required to assist with staging, to optimize results. This study aimed to investigate the accuracy of T stage of Barrett's carcinoma by an AI system based on endoscopic images. METHODS: 1020 images (minimum one per patient, maximum three) from 577 patients with Barrett's adenocarcinoma were used for training and internal validation of a convolutional neural network. In all, 821 images were selected to train the model and 199 images were used for validation. RESULTS: AI recognized Barrett's mucosa without neoplasia with an accuracy of 85 % (95 %CI 82.7-87.1). Mucosal cancer was identified with a sensitivity of 72 % (95 %CI 67.5-76.4), specificity of 64 % (95 %CI 60.0-68.4), and accuracy of 68 % (95 %CI 64.6-70.7). The sensitivity, specificity, and accuracy for early Barrett's neoplasia < T1b sm2 were 57 % (95 %CI 51.8-61.0), 77 % (95 %CI 72.3-80.2), and 67 % (95 %CI 63.4-69.5), respectively. More advanced stages (T3/T4) were diagnosed correctly with a sensitivity of 71 % (95 %CI 65.1-76.7) and specificity of 73 % (95 %CI 69.7-76.5). The overall accuracy was 73 % (95 %CI 69.6-75.5). CONCLUSIONS: The AI system identified esophageal cancer with high accuracy, suggesting its potential to assist endoscopists in clinical decision making.

Terlipressin therapy is associated with increased risk of colonisation with multidrug-resistant bacteria in patients with decompensated cirrhosis.

BACKGROUND: Patients with cirrhosis are susceptible to develop bacterial infections that trigger acute decompensation (AD) and acute-on-chronic liver failure (ACLF). Infections with multidrug-resistant organisms (MDRO) are associated with deleterious outcome. MDRO colonisation frequently proceeds MDRO infections and antibiotic therapy has been associated with MDRO colonisation. AIM: The aim of the study was to assess the influence of non-antibiotic medication contributing to MDRO colonisation. METHODS: Three hundred twenty-four patients with AD and ACLF admitted to the ICU of Frankfurt University Hospital with MDRO screening were included. Regression models were performed to identify drugs associated with MDRO colonisation. Another cohort (n = 129) from Barcelona was included to validate. A third multi-centre cohort (n = 203) with metagenomic sequencing data of stool was included to detect antibiotic resistance genes. RESULTS: A total of 97 patients (30%) were identified to have MDRO colonisation and 35 of them (11%) developed MDRO infection. Patients with MDRO colonisation had significantly higher risk of MDRO infection than those without (p = 0.0098). Apart from antibiotic therapy (odds ratio (OR) 2.91, 95%-confidence interval (CI) 1.82-4.93, p < 0.0001), terlipressin therapy in the previous 14 days was the only independent covariate associated with MDRO colonisation in both cohorts, the overall (OR 9.47, 95%-CI 2.96-30.23, p < 0.0001) and after propensity score matching (OR 5.30, 95%-CI 1.22-23.03, p = 0.011). In the second cohort, prior terlipressin therapy was a risk factor for MDRO colonisation (OR 2.49, 95% CI 0.911-6.823, p = 0.075) and associated with risk of MDRO infection during follow-up (p = 0.017). The validation cohort demonstrated that antibiotic inactivation genes were significantly associated with terlipressin administration (p = 0.001). CONCLUSIONS: Our study reports an increased risk of MDRO colonisation in patients with AD or ACLF, who recently received terlipressin therapy, while other commonly prescribed non-antibiotic co-medications had negligible influence. Future prospective trials are needed to confirm these results.

Bile acid sequestrants in poor healing after endoscopic therapy of Barrett's esophagus.

BACKGROUND/AIMS: Endoscopic therapy for neoplastic Barrett's esophagus (BE) has become the standard of care over the past two decades. In clinical practice, we regularly encounter patients who fail to achieve complete squamous epithelialization of the esophagus. Although the therapeutic strategies in the individual stages of BE, dysplasia, and esophageal adenocarcinoma are well studied and largely standardized, the problem of inadequate healing after endoscopic therapy is only marginally considered. This study aimed to shed light on the variables influencing inadequate wound healing after endoscopic therapy and the effect of bile acid sequestrants (BAS) on healing. METHODS: Retrospective analysis of endoscopically treated neoplastic BE in a single referral center. RESULTS: In 12.1% out of 627 patients, insufficient healing was present 8 to 12 weeks after previous endoscopic therapy. The average follow-up duration was 38.8±18.4 months. Complete healing was achieved in 13 patients already after intensifying proton pump inhibitor therapy. Out of 48 patients under BAS, 29 patients (60.4%) showed complete healing. An additional eight patients (16.7%) improved, but only partial healing was achieved. Eleven (22.9%) patients showed no response to BAS augmented therapy. CONCLUSION: In cases of insufficient healing even under exhaustion of proton pump inhibitors, treatment with BAS can be an option as an ultimate healing attempt.

One Stage Masquelets Technique: Evaluation of Different Forms of Membrane Filling with and without Bone Marrow Mononuclear Cells (BMC) in Large Femoral Bone Defects in Rats.

The classic two-stage masquelet technique is an effective procedure for the treatment of large bone defects. Our group recently showed that one surgery could be saved by using a decellularized dermis membrane (DCD, Epiflex, DIZG). In addition, studies with bone substitute materials for defect filling show that it also appears possible to dispense with the removal of syngeneic cancellous bone (SCB), which is fraught with complications. The focus of this work was to clarify whether the SCB can be replaced by the granular demineralized bone matrix (g-DBM) or fibrous demineralized bone matrix (f-DBM) demineralized bone matrix and whether the colonization of the DCD and/or the DBM defect filling with bone marrow mononuclear cells (BMC) can lead to improved bone healing. In 100 Sprague Dawley rats, a critical femoral bone defect 5 mm in length was stabilized with a plate and then encased in DCD. Subsequently, the defect was filled with SCB (control), g-DBM, or f-DBM, with or without BMC. After 8 weeks, the femurs were harvested and subjected to histological, radiological, and biomechanical analysis. The analyses showed the incipient bony bridging of the defect zone in both groups for g-DBM and f-DBM. Stability and bone formation were not affected compared to the control group. The addition of BMCs showed no further improvement in bone healing. In conclusion, DBM offers a new perspective on defect filling; however, the addition of BMC did not lead to better results.

The Induced Membrane Technique-The Filling Matters: Evaluation of Different Forms of Membrane Filling with and without Bone Marrow Mononuclear Cells (BMC) in Large Femoral Bone Defects in Rats.

The Masquelet technique is used to treat large bone defects; it is a two-stage procedure based on an induced membrane. To improve the induced membrane process, demineralized bone matrix in granular (GDBM) and fibrous form (f-DBM) was tested with and without bone marrow mononuclear cells (BMC) as filling of the membrane against the gold standard filling with syngeneic cancellous bone (SCB). A total of 65 male Sprague-Dawley rats obtained a 5 mm femoral defect. These defects were treated with the induced membrane technique and filled with SCB, GDBM, or f-DBM, with or without BMC. After a healing period of eight weeks, the femurs were harvested and submitted for histological, radiological, and biomechanical analyses. The fracture load in the defect zone was lower compared to SCB in all groups. However, histological analysis showed comparable new bone formation, bone mineral density, and cartilage proportions and vascularization. The results suggest that f-DBM in combination with BMC and the induced membrane technique cannot reproduce the very good results of this material in large, non-membrane coated bone defects, nevertheless it supports the maturation of new bone tissue locally. It can be concluded that BMC should be applied in lower doses and inflammatory cells should be removed from the cell preparation before implantation.

From two stages to one: acceleration of the induced membrane (Masquelet) technique using human acellular dermis for the treatment of non-infectious large bone defects.

INTRODUCTION: The induced membrane technique for the treatment of large bone defects is a two-step procedure. In the first operation, a foreign body membrane is induced around a spacer, then, in the second step, several weeks or months later, the spacer is removed and the Membrane pocket is filled with autologous bone material. Induction of a functional biological membrane might be avoided by initially using a biological membrane. In this study, the effect of a human acellular dermis (hADM, Epiflex, DIZG gGmbH) was evaluated for the treatment of a large (5 mm), plate-stabilised femoral bone defect. MATERIAL AND METHODS: In an established rat model, hADM was compared to the two-stage induced membrane technique and a bone defect without membrane cover. Syngeneous spongiosa from donor animals was used for defect filling in all groups. The group size in each case was n = 5, the induction time of the membrane was 3-4 weeks and the healing time after filling of the defect was 8 weeks. RESULTS: The ultimate loads were increased to levels comparable with native bone in both membrane groups (hADM: 63.2% ± 29.6% of the reference bone, p < 0.05 vs. no membrane, induced membrane: 52.1% ± 25.8% of the reference bone, p < 0.05 vs. no membrane) and were significantly higher than the control group without membrane (21.5%). The membrane groups were radiologically and histologically almost completely bridged by new bone formation, in contrast to the control Group where no closed osseous bridging could be observed. CONCLUSION: The use of the human acellular dermis leads to equivalent healing results in comparison to the two-stage induced membrane technique. This could lead to a shortened therapy duration of large bone defects.

Determination of the effective dose of bone marrow mononuclear cell therapy for bone healing in vivo.

INTRODUCTION: Cell-based therapy by bone marrow mononuclear cells (BMC) in a large-sized bone defect has already shown improved vascularization and new bone formation. First clinical trials are already being conducted. BMC were isolated from bone marrow aspirate and given back to patients in combination with a scaffold within some hours. However, the optimal concentration of BMC has not yet been determined for bone healing. With this study, we want to determine the optimal dosage of the BMC in the bone defect to support bone healing. MATERIAL AND METHODS: Scaffolds with increasing BMC concentrations were inserted into a 5 mm femoral defect, cell concentrations of 2 × 106 BMC/mL, 1 × 107 BMC/mL and 2 × 107 BMC/mL were used. Based on the initial cell number used to colonize the scaffolds, the groups are designated 1 × 106, 5 × 106 and 1 × 107 group. Bone healing was assessed biomechanically, radiologically (µCT), and histologically after 8 weeks healing time. RESULTS: Improved bone healing parameters were noted in the 1 × 106 and 5 × 106 BMC groups. A significantly higher BMD was observed in the 1 × 106 BMC group compared to the other groups. Histologically, a significantly increased bone growth in the defect area was observed in group 5 × 106 BMC. This finding could be supported radiologically. CONCLUSION: It was shown that the effective dose of BMC for bone defect healing ranges from 2 × 106 BMC/mL to 1 × 107 BMC/mL. This concentration range seems to be the therapeutic window for BMC-supported therapy of large bone defects. However, further studies are necessary to clarify the exact BMC-dose dependent mechanisms of bone defect healing and to determine the therapeutically effective range more precisely.