|
Article 1
Early versus on-demand nasoenteric tube feeding in acute pancreatitis. Bakker OJ, van Brunschot S, van Santvoort HC, Besselink MG, Bollen TL, Boermeester MA, Dejong CH, van Goor H, Bosscha K, Ahmed Ali U, Bouwense S, van Grevenstein WM, Heisterkamp J, Houdijk AP, Jansen JM, Karsten TM, Manusama ER, Nieuwenhuijs VB, Schaapherder AF, van der Schelling GP, Schwartz MP, Spanier BW, Tan A, Vecht J, Weusten BL, Witteman BJ, Akkermans LM, Bruno MJ, Dijkgraaf MG, van Ramshorst B, Gooszen HG; Dutch Pancreatitis Study Group. N Engl J Med. 2014 Nov 20;371(21):1983-93.
Nutrition therapy in acute pancreatitis has changed dramatically in the past few decades. Whereas previously the prevailing philosophy espoused “pancreatic rest” with NPO and TPN, current practice is more consistent with “if the gut works, use it.” While the clinical equivalence (or superiority) of enteral over parenteral nutrition is now hardly disputed, the ideal time to initiation and aggressiveness of nutrition delivery remain controversial.
The Dutch Pancreatitis Study Group continues with its streak of performing important research related to the treatment of severe pancreatitis with the PYTHON trial. While previous publications from this group addressed invasive interventions such as endoscopic, video-assisted, and open surgical necrosectomy, this study focused on a more common and less invasive treatment: nutrition therapy. This multicenter study enrolled 208 patients with acute pancreatitis from 19 Dutch hospitals. Included patients were considered at high risk for complications based on APACHE II score (>8), Imrie or modified Glasgow score (>3), or serum C-reactive protein (CRP) level >150 mg/L. Enrolled subjects were then randomly assigned to early nasojejunal tube feeding or “on-demand” oral diet after 72 hours. The early group initiated nutrition a median 23 hours after emergency department (ED) presentation and the on-demand group initiated nutrition a median of 72 hours after ED presentation. Although allowed if requested, only 5% of the on-demand group requested and received food within the first 72 hours. At six months, there was no difference in the primary end point, a composite of major infection or death, leading the authors to conclude non-superiority of early nasojejunal tube feeding.
There are several caveats to consider. First, the sample size was determined based on an expected incidence of the primary composite end point of 40% in the control group. In PYTHON, the actual incidence was only 27%. Thus, this study was underpowered to achieve its stated purpose, but a type II error is relatively unlikely, given that the primary endpoint was actually higher in the study group. Second, while it is true that necrotizing pancreatitis developed in over half of the enrolled patients, on the whole, they were not that sick. Less than 20% of each group required ICU admission and the mean APACHE II score was 11. In many critical care trials, a cut-off of APACHE II score 20 is used to stratify seriously critically ill patients. According to Table 1, only about 30% had respiratory failure (though it is unclear how many required mechanical ventilation) and only about 5% in each group had multiple organ failure. Finally, the control group has come to reflect standard modern practice, which is radically different from standard treatment of severe pancreatitis from even a few years ago (prolonged NPO, TPN). Just as the control arm in the NICE-SUGAR study was different from that of the initial van den Berg, and the control arm of the recently published PROPPR study differed from the initial transfusion ratios from the seminal military studies describing “hemostatic resuscitation”, so, too, this PYTHON trial may be negative because the control arm is already too similar to the study arm thanks to the evolution of standard of care.
In summary, the Dutch PYTHON trial informs us that in patients with acute pancreatitis who are not that sick, either early nasojejunal feeding or oral diet after 72 hours is acceptable, as there was no evidence of benefit or harm associated with either strategy.
Article 2
A Step-up Approach or Open Necrosectomy for Necrotizing Pancreatitis. van Santvoort HC, Besselink MG, Bakker OJ, Hofker HS, Boermeester MA, Dejong CH, van Goor H, Schaapherder AF, van Eijck CH, Bollen TL, van Ramshorst B, Nieuwenhuijs VB, Timmer R, Laméris JS, Kruyt PM, Manusama ER, van der Harst E, van der Schelling GP, Karsten T, Hesselink EJ, van Laarhoven CJ, Rosman C, Bosscha K, de Wit RJ, Houdijk AP, van Leeuwen MS, Buskens E, Gooszen HG; Dutch Pancreatitis Study Group. N Engl J Med. 2010 Apr 22;362(16):1491-502.
Necrotizing pancreatitis complicates 20% of patients presenting with acute pancreatitis. Traditionally treated with open necrosectomy, mortality rates may reach up to 39%. In a hope to limit mortality, less invasive approaches such as percutaneous drainage, endoscopic drainage, and laparoscopic retroperitoneal necrosectomies have been developed. These techniques can be used in a “step-up approach” fashion in order to control the source of infection rather than ensure complete debridement of the necrotic tissue. This approach generally starts with drainage of the infected fluid. If the patient does not improve, then laparoscopic necrosectomy can be performed. Open necrosectomies are reserved for those patients who fail to improve after these measures.
The Dutch Pancreatitis Study Group performed a multi-institutional (19 centers), randomized, controlled trial of 88 patients with confirmed or suspected infected necrotizing pancreatitis between 2005 and 2008. Exclusion criteria included chronic pancreatitis, a previous drainage procedure for pancreatitis (open or minimally invasive), or an acute event requiring laparotomy not for pancreatitis (i.e. visceral perforation, bleeding, or abdominal compartment syndrome). Patients were randomized into open necrosectomy (bilateral subcostal with wide drainage and post-op lavage; n=45) versus the “step-up approach” (n=43). The primary endpoint was a composite of death and major complications (i.e. multisystem organ failure, enterocutaneous fistula, visceral perforation, and hemorrhage requiring intervention) for 3 months after dismissal. Scheduled follow visits were at 3 and 6 months. Power was calculated to require 88 patients in total to detect a difference between 45% and 16% (power = 80%, alpha =0.05). An intention to treat analysis was performed. In the open group, patients underwent a median of 1 necrosectomy (range 1-7) with 42% of patients requiring 2 or more necrosectomies. In addition, 33% required additional percutaneous drainage procedures after abdominal closure. In the “step-up group”, 93% underwent retroperitoneal percutaneous drainage, 2% underwent transabdominal percutaneous drainage, and 5% underwent transgastric drainage. 44% of these patients underwent a second drainage procedure within 72 hours. 35% of the “step-up approach” patients were successfully treated without need for operative drainage. 60%, however, underwent operative necrosectomy 10 days after the initial drainage procedure (range 1-52). The majority of these operative necrosectomies were laparoscopic retroperitoneal approaches (92%) one of which was converted to an open procedure. Two patients (5%) died without an open operation. 33% of these patients required at least one further operative necrosectomy. Regarding the primary composite endpoint of death/major complication, the open group had a significantly greater rate than the “step-up approach” group (69% vs 40%, n=.006; RR 1.75, 95% CI 1.15 - 2.6). In addition, open patients had a greater rate of multiple organ failure (40% vs 12%, p=.001). Death, however, was similar between groups (16% vs 19%, p=.70). Open necrosectomy patients also had greater rates of incisional hernia (24% vs 7%, p=.03), diabetes mellitus (38% vs 16%, p=.02), and enzymatic pancreatic insufficiency (33% vs 7%, p=.002) at 6 months. The “step-up approach” reduced costs by $15,963 per patient. The authors concluded that the “step-up approach” is the preferred treatment strategy for infected necrotizing pancreatitis as it was associated with lower rates of complications, health care utilization, and costs.
Article 3
Laparoscopic transgastric necrosectomy for the management of pancreatic necrosis. Worhunsky DJ, Qadan M, Dua MM, Park WG, Poultsides GA, Norton JA, Visser BC. J Am Coll Surg. 2014 Oct;219(4):735-43.
The surgical group from Stanford evaluated their use of laparoscopic transgastric necrosectomy (LTN) for retrogastric necrotizing pancreatitis. Only patients with a walled-off necrosum (based on duration of pancreatitis and CT findings) and retrogastric location were considered for the procedure. The LTN procedure starts with a standard 12mm umbilical port for laparoscopic evaluation of the abdomen, followed by insufflation of the stomach through esophagogastroscopy. Two additional 5mm ports are placed directly into the stomach (with consideration of an additional stomach port if needed) and then insufflation of the stomach is continued through the gastric ports and endoscope is removed. The pancreatic necrosum is located either by visualization, needle aspiration, or laparoscopic ultrasound and then a wide cystgastrostomy is made. Debridement of the necrosum is performed bluntly and debris is passed through the pylorus. Upon completion of the debridement, the transgastric ports are removed and silk sutures are used to repair the anterior gastrotomy sites. A naso-gastric tube remains to low continuous wall suction for at least 24 hours postoperatively and oral intake resumes based on symptoms and NG output, usually 24 hours postoperatively.
From 2009-2013 the authors had 21 patients who underwent the procedure (62% male, median age 54 years). The cause of pancreatitis was primarily biliary (43%) and alcoholic (33%). The most common indications for the procedure were “persistent unwellness” (no resolution of symptoms) and infected necrosis. Median time between initial presentation and operation was 65 days. Median duration of the LTN procedure itself was 170 minutes with no conversions to laparoscopic or open debridement; two patients required additional image-guided drainage procedures after developing an intra-abdominal fluid collection. Median hospital length of stay was 5 days. Five patients had a major complication related to the procedure (organ failure-14%, GI bleeding-19%) with a 5% mortality rate (1 patient died of major GI bleed from ruptured splenic artery pseudoaneurysm). There were no enterocutaneous or pancreatic fistulae and no incisional hernias. Five patients developed pancreatic insufficiency and 1 had new-onset diabetes.
The discussion portion of this article does a very nice job reviewing the various different techniques available in the operative management of necrotizing pancreatitis. The new technique of LTN offers less morbidity than the open technique but is more successful with the initial procedure than minimally invasive or endoscopic techniques, thus avoiding the need for repeated procedures. An important caveat the authors noted is that several patients had postoperative hemorrhages, underscoring the need for interventional radiology and advanced endoscopy services in hospitals where LTN is performed.
|