Visceral artery pseudoaneurysm in necrotizing pancreatitis: incidence and outcomes.

Background: Visceral artery pseudoaneurysms (VA-PSA) occur in necrotizing pancreatitis; however, little is known about their natural history. This study sought to evaluate the incidence and outcomes of VA-PSA in a large cohort of patients with necrotizing pancreatitis. Methods: Data for patients with necrotizing pancreatitis who were treated between 2005 and 2017 at Indiana University Health University Hospital and who developed a VA-PSA were reviewed to assess incidence, presentation, treatment and outcomes. Results: Twenty-eight of 647 patients with necrotizing pancreatitis (4.3%) developed a VA-PSA between 2005 and 2017. The artery most commonly involved was the splenic artery (36%), followed by the gastroduodenal artery (24%). The most common presenting symptom was bloody drain output (32%), followed by incidental computed tomographic findings (21%). The median time from onset of necrotizing pancreatitis to diagnosis of a VA-PSA was 63.5 days (range 1-957 d). Twenty-five of the 28 patients who developed VA-PSA (89%) were successfully treated with percutaneous angioembolization. Three patients (11%) required surgery: 1 patient rebled following embolization and required operative management, and 2 underwent upfront operative management. The mortality rate attributable to hemorrhage from a VA-PSA in the setting of necrotizing pancreatitis was 14% (4 of 28 patients). Conclusion: In this study, VA-PSA occurred in 4.3% of patients with necrotizing pancreatitis. Percutaneous angioembolization effectively treated most cases; however, mortality from VA-PSA was high (14%). A high degree of clinical suspicion remains critical for early diagnosis of this potentially fatal problem.

Contexte: Les faux anévrismes des artères viscérales (FAAV) surviennent en présence d'une pancréatite nécrosante; on en sait cependant peu sur leur histoire naturelle. L'objectif de l'étude était d'évaluer l'incidence et les issues des FAAV dans une grande cohorte de patients atteints de pancréatite nécrosante. Méthodes: Nous avons examiné les données des patients atteints de pancréatite nécrosante traités entre 2005 et 2017 à l'Hôpital universitaire de l'Université de l'Indiana qui ont fait un FAAV afin d'évaluer l'incidence, les premiers signes, le traitement et les issues de cette affection. Résultats: Vingt-huit (4,3 %) des 647 patients atteints de pancréatite nécrosante inclus (2005­2017) ont fait un FAAV. L'artère la plus souvent touchée était l'artère splénique (36 %), suivie de l'artère gastroduodénale (24 %). Les premiers signes les plus courants étaient la présence de sang dans les liquides évacués par drainage (32 %), puis les résultats d'une tomodensitométrie effectuée pour une autre raison (21 %). Le délai médian entre l'apparition de la pancréatite nécrosante et le diagnostic de FAAV était de 63,5 jours (intervalle : 1 à 957 jours). Vingt-cinq des 28 patients ayant fait un FAAV (89 %) ont été traités avec succès par angioembolisation percutanée. Trois patients (11 %) ont dû être opérés : 2 dès le début, et le troisième parce qu'il a recommencé à saigner après l'embolisation. Le taux de mortalité par hémorragie due à un FAAV chez les personnes atteintes d'une pancréatite nécrosante était de 14 % (4 patients sur 28). Conclusion: Dans cette étude, 4,3 % des patients atteints de pancréatite nécrosante ont connu un FAAV. L'angioembolisation percutanée s'est avérée efficace dans la plupart des cas; cependant, la mortalité associée aux FAAV était élevée (14 %). Il est crucial de faire preuve d'une grande suspicion clinique afin de diagnostiquer tôt cette affection potentiellement mortelle.

Lung biopsy in children's interstitial and diffuse lung disease: Does it alter management?

INTRODUCTION: Pediatric patients with acute life-threatening consequences of interstitial and diffuse lung disease are often treated with empiric systemic corticosteroids, immune modulators, and/or broad antibiotic therapy. Histological evaluation of lung tissue represents the final necessary step in diagnosis-however, a definitive diagnosis may still remain elusive and medical therapies may not be changed following biopsy. We hypothesized that lung biopsy from pediatric patients with children's interstitial and diffuse lung disease (chILD) without a defined lesion on computed tomography (CT) imaging would guide diagnosis, but not substantially alter clinical management. METHODS: After IRB approval, patients who underwent a lung biopsy at a single large children's hospital between 2013 and 2018 were retrospectively reviewed. Patients without a defined lesion were included. Demographics, length of stay, oxygen-requirements, steroid, unique number of immune modulators, and antibiotics prebiopsy and postbiopsy were reviewed. Nonparametric data were compared by the Mann Whitney U and Kruskal Wallace tests and expressed as median with interquartile range. Decision tree alterations were analyzed by t test. P < .05 was significant. RESULTS: Sixty-four patients underwent lung biopsy during the period. Nineteen (30%) did not have a defined lesion on CT scan, and were included. A significant difference was seen between prebiopsy, 2 weeks, and 2 months postbiopsy prednisone dosing (P = .03), while the number of unique immune modulators, antibiotics, type of oxygen support and FiO2 were not significantly different before or after obtaining biopsy results. Pathology results provided additional information in 12 of 19 (63%) patients which resulted in management changes. CONCLUSIONS: Lung biopsy in chILD may guide clinical management, especially influencing the management of steroid dosing. Although on aggregate the number of antibiotics, immune modulators, mode of oxygen support and FiO2 did not differ significantly before and after biopsy, the pathologic evaluation provided diagnostic information that led to a variety of changes in therapeutic management in greater than half of the population.

Pancreatic Adenocarcinoma Causing Necrotizing Pancreatitis: Not as Rare as You Think?

BACKGROUND: Necrotizing pancreatitis (NP) presents a unique clinical challenge because of its complex and lengthy disease course. Pancreatic necrosis occurs in 10%-20% of acute pancreatitis cases and may result from any etiology. Scattered reports describe pancreatic tumors causing NP; however, the relationship between these disease processes is not clear. We have treated patients whose NP was caused by pancreatic ductal adenocarcinoma (PDAC) and therefore sought to clarify the clinical outcomes of these patients. METHODS: Patients treated between 2005 and 2018 for NP caused by PDAC were identified. The relationship between NP and PDAC was examined, and the clinical courses of both disease processes were evaluated. RESULTS: Among 647 patients treated for NP, seven patients (1.1%) had PDAC and NP. The mean age at NP diagnosis was 60.6 y (range, 49-66). Two patients had postprocedural pancreatitis after cancer diagnosis, and the remaining five patients had NP caused by PDAC. Median duration between diagnoses of NP and PDAC was 5.6 mo (range, 3.5-21.8). For PDAC treatment, four patients received chemotherapy alone, one received palliative radiation therapy, and one died without oncologic management. One patient underwent operative resection of PDAC. Median survival was 12.7 mo (range, 0.4-49.9). CONCLUSIONS: PDAC may be a more common cause of NP than previously considered and should be considered in patients with NP of appropriate age in whom etiology is otherwise unclear. Prompt diagnosis facilitates optimal treatment in this challenging clinical situation.

Disconnected Pancreatic Duct Syndrome: Spectrum of Operative Management.

BACKGROUND: Disconnected pancreatic duct syndrome (DPDS) is common after necrotizing pancreatitis (NP). Surgical management may be by internal drainage or left (distal) pancreatectomy. Therapeutic decision-making must consider sinistral portal hypertension, parenchymal volume of disconnected pancreas, and timing relative to definitive management of pancreatic necrosis. The aim of this study is to evaluate outcomes after operative management for DPDS. METHODS: All patients with NP undergoing an operation for DPDS were included in the study (2005-2017). Perioperative outcomes and long-term durability were evaluated. RESULTS: Among 647 patients with NP, 299 (46%) had DPDS. Operative management was required in 202/299 (68%) patients with DPDS. Median follow-up was 30 mo (2-165). Definitive operative therapy included internal drainage (n = 111) or resection (n = 91). Time from NP diagnosis to operation was 126 d (20 d to 81 mo). Overall morbidity was 46%. Postoperative length of stay was 7 d (2-97). Readmission was required in 39 patients (19%). Mortality was 2%. Repeat pancreatic intervention was required in 23 patients (11%) at a median of 15 mo (1-98). Repeat pancreatectomy was performed in nine patients and the remaining 14 patients were managed with endoscopic therapy. CONCLUSIONS: DPDS is a common and challenging consequence of NP. Appropriate operation is durable in nearly 90% of patients.

High Rates of Readmission in Necrotizing Pancreatitis: Natural History or Opportunity for Improvement?

BACKGROUND: Necrotizing pancreatitis (NP) is a complex and heterogeneous disease with a protracted disease course. Hospital readmission is extremely common; however, few data exist regarding the cause of readmission in NP. METHODS: A retrospective review of NP patients treated between 2005 and 2017 identified patients readmitted both locally and to our hospital. All patients with unplanned hospital readmissions were evaluated to determine the cause for readmission. Clinical and demographic factors of all patients were recorded. As appropriate, two independent group t tests and Pearson's correlation or Fisher's exact tests were performed to analyze the relationship between index admission clinical factors and readmission. p values of < 0.05 were accepted as statistically significant. RESULTS: Six hundred one NP patients were reviewed. Median age was 52 years (13-96). Median index admission length of stay was 19 days (2-176). The most common etiology was biliary (49.9%) followed by alcohol (20.0%). Unplanned readmission occurred in 432 patients (72%) accounting for a total of 971 unique readmissions (mean readmissions/patient, 2.3). The most common readmission indications were symptomatic necrosis requiring supportive care and/or intervention (31.2%), infected necrosis requiring antibiotics and/or intervention (26.6%), failure to thrive (9.7%), and non-necrosis infection (6.6%). Patients requiring readmission had increased incidence of index admission renal failure (21.3% vs. 14.2%, p = 0.05) and cardiovascular failure (12.5% vs. 4.7%, p = 0.01). DISCUSSION: Readmission in NP is extremely common. Significant portions of readmissions are a result of the disease natural history; however, a percentage of readmissions appear to be preventable. Patients with organ failure are at increased risk for unplanned readmission and will benefit from close follow-up.