Acute pancreatitis associated with pancreatic necrosis is associated with a severe
and protracted disease course. Acute necrotizing pancreatitis (ANP) in the initial
phase is associated with acute necrotic collections (ANC) and this over a period of
4 weeks or more gets liquefied and walled off leading on to formation of walled off
necrosis (WON).[1] The WON have a variable clinical course with some spontaneously resolving and some
getting symptomatic or infected. These symptomatic WON needs to be treated and the
majority of these are to be drained. Simple catheter drainage is not effective for
these necrotic collections as they contain varying amount of solid necrotic debris
that cannot be removed through small caliber catheters. These collections were earlier
treated with surgical necrosectomy but over past decade there has been paradigm shift
in their management toward minimally invasive interventional radiologic, laparoscopic,
and endoscopic procedures.[2]
Endoscopic transmural drainage is the preferred drainage route for WON. However, the
initial results endoscopic transmural drainage with a single plastic stent was disappointing
because of ineffective drainage of solid necrotic material by thin caliber plastic
stent. However, with use of more aggressive endoscopic drainage techniques such as
larger tract dilation, placement of multiple stents, insertion of nasocystic catheter
(NCC) along with aggressive irrigation, use of fully covered self-expanding metallic
stents (FCSEMS)/lumen apposing metal stents, and direct debridement of solid necrotic
tissue by endoscopic necrosectomy has led on to considerable improvement in the results
of endotherapy.[3],[4],[5],[6],[7],[8]
Out of all the endoscopic procedures, direct endoscopic necrosectomy (DEN) has been
reported to be most effective as it involves direct removal of necrotic tissue and
therefore leads on to quicker resolution.[6] However, DEN, is associated with increased risk of complications such as perforation
and bleeding and is also a time-consuming procedure. Moreover, lack of specific necrosectomy
accessories makes it a difficult procedure that can be done by experts only. Furthermore,
there is evidence in the literature that majority of patients with WON can be treated
with large balloon dilatation and multiple plastic stents.[4],[5],[7] Therefore, a “step-up” endoscopic approach has been advised for treatment for WON
where DEN is reserved for patients not responding to drainage and irrigation. This
approach is similar to step-up approach used for ANC that involves an initial percutaneous
drainage (PCD) followed by endoscopic or surgical or laparoscopic necrosectomy if
required.[9],[10] In this news and views, we will discuss two different step-up endoscopic approaches
used for treatment of WON at two different tertiary care centers in India and published
recently.
First by Lakhtakia et al. from Asian Institute of Gastroenterology, reported the efficacy
of endoscopic step-up approach using a dedicated covered biflanged metal stent (BFMS)
in 205 patients of WON.[11] Second study by Rana et al. from Postgraduate Institute of Medical Education and
Research reported the efficacy of endoscopic step-up approach using multiple plastic
stents in 84 patients.[12]
Lakhtakia et al. reported a retrospective study wherein consecutive patients with
symptomatic WON seen over a 3-year period were treated with BFMS. All the patients
were treated with endoscopic ultrasound (EUS)-guided transmural drainage. The EUS-guided
drainage procedure was performed in a standard manner wherein the WON was punctured
using a 19-gauge needle followed by insertion of guide wire. The transmural tract
was dilated using a 6F cystotome followed by 6-mm balloon dilator. Thereafter, the
stent delivery catheter of BFMS (Nagi; Taewoong Medical, Gyeonggi-do, South Korea)
was advanced over the guidewire across the WON wall and deployed. After the drainage,
patients were admitted and observed and in patients with persistent or new onset symptoms,
a noncontrast computed tomography (CT) was done CT to assess response and diagnose
any adverse events. In nonresponders, endoscopic assessment of BFMS was done to look
for stent clogging by the necrotic material. As first step in this step-up approach,
any occluding debris was cleared using endoscopic devices such as snare or forceps.
Patients not responding with de-clogging, underwent placement of a nasocystic tube
(NCT) through the lumen of the BFMS. The WON cavity was intermittently irrigated with
3% hydrogen peroxide (H2O2), and saline solution with each session of irrigation being
done at 8 hourly intervals using 20 ml of H2O2 followed 10 min later by 100 ml saline
solution. Patients still not responding underwent DEN using gastroscope (Olympus GIF
XTQ160 or GIF HQ190) with water jet facility through the BFMS using accessories such
as braided snares, forceps, nets, and stone retrieval baskets.
Rana et al. reported a different step-up approach wherein the EUS-guided drainage
procedure is almost similar to the procedure described above except that the transmural
tract was dilated using an endoscopic retrograde cholangiopancreatography cannula
or 4 mm biliary balloon dilator instead of a cystotome. Moreover, if it was not possible
to dilate the tract with noncautery methods because of thickened wall, it was dilated
using a wire-guided cautery needle knife. Thereafter, the tract was further dilated
up to 8–15 mm with a wire-guided hydrostatic balloon. Also in this study, the WON
were assessed for the quantity of the necrotic material present. The echogenic material
present in the WON on EUS was suggestive of solid debris. The quantification of necrotic
debris was an approximate visual judgment of the endoscopist and mean of the findings
of two experienced endosonologists who separately reviewed the EUS images to quantify
the solid debris was taken as the amount of solid debris in WON. Following balloon
dilatation, 1–3 7-Fr double-pigtail stents, between 3 and 7 cm in length, were inserted
into WON. Moreover, in patients with >10% solid debris in walled off pancreatic necrosis
(WOPN), a 7-Fr NCC was also positioned alongside stents for irrigation with 200 ml
of normal saline every 4–6 hourly. No H2O2 was used in this study. All patients underwent
CT of the abdomen 72 h after drainage and NCC was removed in patients who had symptomatic
improvement with> 50% reduction in size of WON. Patients with new onset fever or worsening
of existing symptoms underwent repeat endoscopic transmural drainage with further
dilatation of tract up to 18 mm in stomach and 12 mm in duodenum) and multiple 10
Fr stents, 3–5 cm in length were inserted. However, if after second or subsequent
session of drainage, WON persisted with persisting symptoms, a decision for additional
transmural drainage by stents, FCSEMS insertion, DEN or surgery was taken after reviewing
imaging findings, discussion with patient and his/her family, and consultation with
pancreatic surgeons.
Lakhtakia et al. studied 205 patients (mean age: 34.8 ± 12.8 years; 81% males) with
WON who underwent EUS-guided drainage with BFMS placement. The mean size of WON was
10.8 cm and mean duration between onset of pancreatitis and drainage was 42 ± 12.5
days. The procedure was technically successful in all patients with stent deployment
requiring a second procedure in 2 patients. Procedure related adverse events were
observed in 8 (3.9%) patients with 4 patients having self-limiting bleeding, 2 having
major bleeding with one requiring angiographic coil embolization and another patient
requiring surgery. Furthermore, two patients developed perforation requiring surgery.
In the study group, reintervention was required in 49 (23.9%) patients with stent
dysfunction due to occluding debris observed in 21 patients of which 10 patients (4.8%)
improved with de-clogging only. The other 11 patients required NCT placement with
irrigation due to the presence of solid debris within the WON. Twenty eight patients
directly underwent NCT placement and thus 39 patients required second step of NCT
placement. NCT was maintained for 3–7 days and led on to resolution of WON in 10 patients.
Six patients with adherent mild debris required placement of a plastic stent within
the BFMS after removal of NCT. Remaining 23 patients (11%) required DEN with 12 patients
requiring one session, six needing two sessions, four needing three sessions, and
one needing four sessions of DEN. Four patients had treatment failure with two requiring
surgery and two 2 needing additional PCD for WON extending to the left paracolic gutter.
Migration of BFMS occurred in 5 (2.4%) patients during DEN and these stents could
be replaced in all patients.
Of 205 patients, clinical success was achieved in 153 patients (74.6%) by step one
of drainage with BFMS. Of the remaining 49 patients, 10 patients improved with de-clogging,
16 patients with NCT placement and 19 patients with DEN thereby yielding an overall
treatment success rate of 96.5% patients. On follow-up, 162 (79%) patients underwent
uncomplicated removal of BFMS within 4–8 weeks of placement. However, 21 patients
reported late (between 3 and 12 months after placement) due to logistical reasons
and removal was uncomplicated in 12 patients and nine patients required Argon plasma
coagulation (APC) of the gastric wall around the BFMS with or without its piecemeal
removal. Symptomatic pancreatic fluid collection (PFC) recurrence occurred in five
patients and all of them patients had disconnected pancreatic duct. The recurrence
was treated successfully by EUS-guided drainage using plastic stents.
Rana et al. studied 86 patients (73 [84.9%] males) and mean age 38.24 ± 10.11 years
with symptomatic WON. The patients presented after a mean duration of 10.8 ± 2.8 weeks
of onset of acute attack of ANP and mean size of WON was 10.7 ± 2.9 cm. EUS-guided
transmural drainage was technically successful in 85 (98.8%) patients with no collateral-free
window being identified in one patient with splenic vein thrombosis and extensive
intra-abdominal collaterals. Among 85 patients who underwent endoscopic drainage,
number of endoscopic procedures needed was one in 5 (6%) patients, 2 in 25 (29.5%),
3 in 30 (35.3%), 4 in 16 (18.7%) patients and more than 4 procedures in 9 (10.5%)
patients. 78/85 (92%) patients underwent a NCC catheter placement at the time of initial
EUS-guided drainage for a period of 3–10 days. Seventy (82.4%) were successfully drained
using multiple 7/10 Fr plastic stents alone while DEN was needed in 9 (10.6%) and
FCSEMS placement in 6 (7%) patients. Of six patients in whom FCSEMS was placed, 5
(84%) required additional DEN through the metallic stent.
None of the patients required surgery but 2 (2.3%) patients' required additional PCD
because of presence of undrained peripheral WON. The time to resolution of WON was
4.6 ± 1.3 weeks. The complications noted were self-limiting pneumoperitoneum because
of puncture of WON cavity while performing DEN (n = 1), bile leak following cholecystecomy (n = 1), development of external pancreatic fistula following PCD (n = 1) and gastrointestinal bleeding requiring blood transfusion (n = 1). On comparison of patients who were successfully treated with multiple plastic
stents versus patients who required SEMS/DEN, it was found that the patients who needed
DEN/FCSEMS presented earlier and had large size collection with more solid necrotic
debris as compared to patients treated with multiple plastic stents alone. The mean
amount of solid debris in patients successfully treated with multiple plastic stents
was 26%, and this was significantly less than the amount of solid debris in patients
treated with DEN/FCSEMS (47%; P< 0.001).
COMMENTARY
WON is a heterogeneous group of PFC's with varying amount of solid necrotic debris,
and therefore one procedure may not be effective and safe in all the patients. The
results of the above-mentioned two studies reemphasize the fact that majority of patients
with WON can be successfully treated endoscopically using multiple plastic stents
or FCSEMS without resorting to labor intensive and complicated DEN. In patients with
infected pancreatic necrosis, a “step-up” approach with minimally invasive PCD followed
by necrosectomy in nonresponders has been advocated.[9],[10] This step-up approach avoids invasive necrosectomy and associated complications
in almost one-third of patients. The studies by Rana et al. and Lakhtakia et al. have
also shown that using an endoscopic step-up approach, 82%–89% of symptomatic patients
with WON could be successfully treated with drainage techniques only and thus avoiding
DEN. Bang et al. also reported that using a tailored algorithmic approach of multiple
plastic stents/NCT and/or multiple transluminal gateway techniques based on size and
extent of WOPN and stepwise response to intervention, 91% patients with WOPN will
have successful outcome.[13] Rana et al. also reported that WON having large size and more solid debris needed
more aggressive therapeutic method for successful outcome with patients having> 40%
solid debris either needing DEN or surgical necrosectomy.[8] Out of the two step-up approaches, discussed here, which one is better can be answered
by comparative studies alone. Moreover, whenever, there is more than one way to skin
the cat, you should choose the one in which you are the best. We had earlier proposed
a classification system for WON base on the amount of solid necrotic debris with WON
having [14] WON with more than 40% was labeled as walled off solid necrosis and these patients
usually require FCSEMS with or without DEN.
