Acute pancreatitis (AP), especially acute necrotizing pancreatitis, is a potentially
life-threatening condition with high morbidity and mortality.[1] Peri-pancreatic fluid collections (PFCs) are one of the common local complications
of AP and according to the revised Atlanta classification, they have been classified
into four types depending on the content and presence or absence of well-formed wall.[1] Walled of necrosis (WON) develops in the delayed phase of illness in patients with
acute necrotizing pancreatitis. It is an encapsulated collection of fluid and solid
necrotic material that usually develops after 4 weeks of an attack of acute necrotizing
pancreatitis.[1] Majority of PFCs including asymptomatic WON resolve spontaneously without any need
of intervention.[2],[3] However, in the presence of symptoms such as persistent significant pain, infection,
gastric outlet obstruction, or biliary obstruction, the PFC requires drainage and
it can be achieved using radiological, endoscopic, or surgical means.
The management algorithm of symptomatic PFCs has evolved from primarily invasive surgical
necrosectomy to minimally invasive and now endoscopic step-up approach.[4]
[5]
[6] A multicentric randomized trial comparing open necrosectomy versus minimally invasive
step-up approach has shown decreased risk of new-onset organ failure, new-onset diabetes
mellitus, or death in minimally invasive step-up approach, comprising initial percutaneous
catheter drainage followed by laparoscopic or surgical necrosectomy, if required.[6] Recently, endoscopic step-up approach has also shown a technical success rate of
98%–99% and a clinical success rate up to 95%.[4],[5]
The face of endoscopic step-up approach has changed in the last decade due to the
introduction of lumen-apposing metal stents (LAMS) such as Axios stent (Boston Scientific,
Natick, MA, USA) and Niti-S Spaxus stent (Taewoong Medical, Gyeonggi-do, South Korea).
Recently, the introduction of electrocautery-enhanced delivery system has made insertion
of Axios as well as Spaxus stent even simpler with minimal technical complications.
As WON is a mixture of solid as well as liquid necrotic materials, LAMS are preferred
as their large diameter is expected to be beneficial for effective drainage of solid
necrotic material and also, if required, allow easy direct endoscopic necrosectomy
(DEN) without the need of repeated large diameter balloon dilatations.
In a recent retrospective comparison between LAMS and double-pigtail plastic stents
(DPPSs) in 94 patients, the authors have shown that WON resolution rate without DEN
was higher with LAMS compared to DPPSs (60.4% vs. 30.8%; P = 0.01), with lower need for repeated necrosectomy and decreased incidence of intervention-related
hemorrhage.[7] Despite few disadvantages of LAMS such as higher cost and stent blockade by necrotic
debris requiring repeated endoscopic interventions, it appeared that we have finally
developed an ideal stent for the management of WON. However, recent concerns about
the serious adverse effects because of prolonged duration of transmural stenting by
LAMS have dampened this initial enthusiasm with LAMS. In this news and views, we will
discuss two recent studies on the endoscopic management of WON with LAMS and have
highlighted LAMS-related serious complications. Bang et al. conducted a prospective randomized trial comparing LAMS and DPPSs in patients with
WON,[8] whereas Brimhall et al. conducted a retrospective study comparing LAMS and DPPSs in patients with WON.[9]
Bang et al. enrolled sixty patients of WON prospectively and randomized them to endoscopic ultrasound-guided
drainage using LAMS or plastic stents. The patients had symptomatic WON either in
the form of infection or persistent pain requiring narcotics at least three times
per day. All patients underwent transmural drainage using either LAMS (Hot Axios (Boston
Scientific, Natick, Massachusetts, USA ); 15 mm in diameter and 10 mm in length) or
DPPSs (7F 4 cm). In patients with WON >120 mm in size, multigate technique was used
using the same stent type. In addition, nasocystic catheters (10 Fr) were inserted
for lavage of necrotic cavity. Patients with inadequate response as defined by persistent
or new-onset systemic inflammatory response syndrome (SIRS) underwent computed tomography
(CT) scan at 72 h postintervention. If CT showed <25% decrease in WON size, then the
patient underwent creation of additional transmural tract using the same stent technique
or DEN depending on the content of WON. Patients were followed up at 6 weeks postintervention
with CT scan to see for WON size and endoscopic retrograde cholangiopancreaticogram
to look for pancreatic duct integrity. In the presence of resolution of WON with intact
main pancreatic duct, transmural stents were removed. In the presence of disconnected
pancreatic duct syndrome (DPDS) with resolution of WON, plastic stents were left in situ indefinitely and LAMS were replaced with plastic stents. In the presence of partial
disruption, transpapillary stent was placed. Treatment success was defined as resolution
of WON on CT scan with resolution of symptoms at 6 months. Treatment failure was defined
as need of rescue surgery or death due to disease or intervention.
There were 31 patients in the LAMS group and 29 patients in the plastic stent group.
Both the groups were identical in baseline characteristics including a mean size of
WON (10.2 ± 4.6 vs. 10.7 ± 6.8 cm; P = 0.784). In the LAMS group, 29% of patients had ongoing SIRS or organ failure, and
the median degree of necrosis was 40% (interquartile range [IQR] 20) prior to intervention.
Multigate technique was used in 9.7% of patients, and a mean of 2.8 procedures were
required to achieve a clinical success rate of 93.5%. In view of inadequate response
at 72 h, two patients underwent percutaneous drainage catheter placement, two patients
underwent creation of additional transmural track, and four patients underwent DEN.
Stent-related adverse events occurred in ten (32.2%) patients. Two patients presented
with stent buried in the gastric wall and three patients presented with massive gastrointestinal
(GI) bleed requiring intensive care unit admission and blood transfusion. In all the
three patients with bleed, endoscopic ultrasonography examination showed the presence
of interlacing vessels within the distal flange of LAMS. CT angiography followed by
glue embolization of pseudoaneurysm was required in all the three patients. Three
patients presented with obstructive jaundice due to mechanical obstruction by LAMS
when it was deployed via the duodenal bulb. All these adverse events were observed
in the LAMS cohort at 3 or more weeks postprocedure. Hence, the study protocol was
changed as per the safety board recommendation, and CT scan was obtained at 3 weeks
followed by stent removal if there was resolution of WON. After these changes, only
two patients had stent-related adverse events: one patient presented with stent migration
in GI tract with obstructive symptoms and another patient had bled from the gastric
mucosa while using electrocautery-enhanced delivery system. Pancreatography showed
normal duct in 9 patients, DPDS in 13 patients, partial duct disruption in 6 patients,
and duct status was unknown in 3 patients. Out of the 13 patients with DPDS, transmural
plastic stent exchange was not possible in six patients due to collapse of the necrotic
cavity. All patients with partial leak were treated with transpapillary stent bridging
the leak.
In plastic stent group, ongoing SIRS was present in 44.8% of patients, and the median
degree of necrosis was 50% (IQR 20). Multigate technique was used in 31% of patients,
and a mean of 3.2 procedures were required to achieve a treatment success rate of
96.6%. In view of inadequate response at 72 h, five patients underwent percutaneous
drainage catheter placement, five patients underwent creation of additional transmural
track, and six patients underwent DEN. Stent-related adverse events occurred in two
patients, both presented with migration of stents in the jejunum which were retrieved
using rat tooth forceps. Pancreatography revealed normal pancreatic duct in 9 patients,
DPDS in 17 patients, partial duct disruption in 2 patients, and ductal status was
unknown in 1 patient. Both patients with partial disruption were treated with bridging
transpapillary stents, and the stents were left in situ in patients with DPDS.
At 6 months postprocedure, both groups (LAMS and plastic groups) had similar clinical
success (93.5% vs. 96.6%; P = 0.99), length of hospital stay (6.2 ± 9.0 vs. 12.2 ± 21.1 days; P = 0.129), and total number of procedures required (2.8 ± 1.2 vs. 3.2 ± 1.5; P = 0.192). Although overall adverse events were similar in both the groups (P = 0.077), stent-related adverse events were more frequent in the LAMS group (32.3%
vs. 6.9%; P = 0.014). However, stent-related adverse events were more frequent in the original
protocol (P = 0.005), and after change in protocol at 5 months, stent-related adverse event rate
was similar in both the groups (P = 0.999). The duration of procedure was statistically significantly less in LAMS
group (15 vs. 40 min; P < 0.001), and the cost of procedure was significantly more in LAMS group (12,155
vs. 6609 USD; P < 0.001).
The authors hypothesized that, due to wider diameter, WON resolved rapidly in the
LAMS group. By virtue of its lumen-opposing nature, once the necrotic cavity collapses,
it gets embedded deeply into the wall or causes compression of the surrounding structures
including bile duct or vessels causing obstructive jaundice or pseudoaneurysm, respectively.
Moreover, these complications occurred in patients after 3 weeks of placement of LAMS.
Hence, the authors recommended doing CT scan at 3 weeks postprocedure and remove LAMS
if WON has resolved to reduce stent-related complications. The authors concluded that,
except for procedure duration, there was no significant difference in the clinical
outcomes of patients with WON treated with LAMS or plastic stents.
Brimhall et al. conducted a retrospective study of 249 patients with symptomatic pancreatic pseudocyst
or WON: 97 (39%) patients in the LAMS group (10 or 15 mm diameter of LAMS) and 152
(61%) patients in the DPPS group (at least two 7/10F plastic stents).[9] Baseline characteristics including the presence of WON were similar in both the
groups (83.3% vs.Shah and Rana: LAMS related complications 76.3%; P = 0.52). However, the LAMS group had larger size of fluid collection (80.1 vs. 69.8
mm; P = 0.001), more patients with >50% necrosis in WON (56.7% vs. 36.3%; P = 0.038), and less patients with chronic pancreatitis (20.6% vs. 38.8%; P = 0.03). In the LAMS group, eight patients had DPPSs through LAMS and in the DPPSs
group, patients had a median number of two stents placed. Technical success was similar
in the LAMS and DPPS groups (90.1% vs. 92.8%; P = 0.67). Clinical success (90.1% vs. 91.8%; P = 0.54), average number of endoscopic necrosectomies (1.7, range 1–11 vs. 1.9, range
1–19; P = 0.93), and stent duration (51.5 vs. 59.8 days; P = 0.26) were also similar in both the groups. The authors have also stated that location
of fluid collection, overall presence of necrosis, or use of nasocystic drain were
not significant factors contributing to technical or clinical success or adverse event
occurrence.
Overall adverse events were similar in the LAMS and DPPS groups (24.7% vs. 17.8%;
P = 0.67). However, in the LAMS group, there was presence of significantly higher rate
of bleeding episodes (15.46% vs. 3.28%; P = 0.0005) and higher rate of pseudoaneurysmal bleeding (8.2% vs. 0.7%; P = 0.009). Overall average time to bleed was 18.5 postprocedure days. Infection (2.0%
vs. 3.9%; P = 0.28) and perforation rates (0% vs. 2%; P = 0.22) were showing higher trend in the DPPS group.
The authors hypothesized that the presence of large peripancreatic fluid collection
might compress the vessel, making it more liable to be missed during predrainage cross-sectional
imaging. After drainage due to rapid emptying of the cavity size, distal flanges may
irritate the cyst wall, promoting pseudoaneurysmal formation. However, though both
groups had similar clinical efficacy, the LAMS group had larger collection size with
more necrosis content in WON compared to the DPPS group. The authors in this study
concluded that LAMS should be used in the presence of collection with larger diameter
and more necrosis content as subsequent drainage and endoscopic necrosectomy, if required,
can be more feasible. Once there is resolution of collection cavity, LAMS should be
removed to reduce the chances of bleeding.
Commentary
The management of pancreatic collection, especially WON, is a challenging task in
the management of AP. Patients with WON have variable natural history of progression,
with some WON becoming symptomatic, some getting resolved spontaneously, and some
getting infected.[10] Moreover, treatment of such symptomatic WON is also a challenging task due to the
presence of variable amount of necrotic debris which is difficult to remove by traditional
endoscopic or percutaneous approach. The introduction of LAMS was a game changer in
the endoscopic management of WON as along with being easier to deploy, it provides
wider diameter for drainage of solid necrotic content as well as makes DEN easier,
if required. A meta-analysis has shown technical success rate up to 98% and clinical
success rate up to 93% in WON using LAMS.[11] However, the use of LAMS is associated with higher cost of procedure as well as
various side effects such as bleeding, stent clogging by necrotic debris, and stent
migration. Bang et al. prospectively compared LAMS with plastic stents and showed that both modalities
are associated with similar clinical success rate at 6 months.[10] However, the LAMS group is associated with significantly higher risk of pseudoaneurysm-related
bleeding, especially if the stent is kept in situ for more than 3 weeks. Brimhall et al. in their retrospective study also showed that both types of drainage methods have
similar technical and clinical success rates.[9] In this study also, the LAMS group had higher rate of pseudoaneurysm-related bleeding.
In the study by Bang et al., the authors suggested that CT scan should be performed at 3 weeks and if it shows
resolution of cavity, the LAMS should be removed. The authors have demonstrated that
this protocol of removing LAMS at 3 weeks decreased the risk of pseudoaneurysm-related
bleeding.[10] Recently, a study by Rana et al. suggested a hybrid approach of early removal of metal stents followed by placement
of multiple plastic stents in WON. In their study, after placement of fully covered
self-expanding metal stent, they have exchanged the clogged SEMS early within 2 weeks
(median 10.9 ± 1.3 days) with multiple 10F plastic stents (median of 3). They have
compared such hybrid approach (n = 10) with the conventional endoscopic de-clogging of SEMS and repeated endoscopic
necrosectomy approach (n = 17). They found that patients with hybrid approach had faster resolution of collection
(20.3 ± 3.1 days vs. 25.6 ± 5.5 days; P = 0.01) and required fewer endoscopic procedures (2.6 ± 0.5 vs. 4.1 ± 1.1; P = 0.0003). They have hypothesized that early removal of such clogged stent leads
to removal of large chunk of adherent necrotic debris, leading to early resolution
of collection and thus also avoids the risks of delayed removal of metal stents.[12]
The current evidence suggests that there is probably no significant difference in
clinical as well as technical outcome between patients with WON treated with LAMS
versus plastic stents. However, LAMS is technically easier to insert with a shorter
procedure time as compared to plastic stents and also provides easy access for DEN,
whenever required. The decision of placement of LAMS should be based on larger size
of collection with higher percentage of necrotic debris, presence or absence of large
vessel or pseudoaneurysm in the vicinity of collection, and financial status of the
patient. Given the concerns of bleeding and other complications, LAMS should be removed
early, preferably within 3 weeks of insertion.[13] It seems that LAMS are still not an ideal stent for endoscopic treatment of WON
and therefore, the search for an ideal stent continues.
