Gastroenterology 2018;-:1–6

1 61
2 62
3 63
4
5
American Gastroenterological Association Institute Guideline 64
65
6 on Initial Management of Acute Pancreatitis 66
7 67
8 Q3 Seth D. Crockett,1 Sachin Wani,2 Timothy B. Gardner,3 Yngve Falck-Ytter,4,5 and Alan Barkun6; 68
9 on behalf of American Gastroenterological Association Institute Clinical Guidelines Committee 69
10 70
11 1
Division of Gastroenterology and Hepatology, University of North Carolina School of Medicine, Chapel Hill, North Carolina; 71
12 2
Division of Gastroenterology and Hepatology, University of Colorado, Anschutz Medical Campus, Aurora, Colorado; 72
13 3
Section of Gastroenterology and Hepatology, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire; 4Division of 73
14 Gastroenterology, Case Western Reserve University, Cleveland, Ohio; 5Louis Stokes VA Medical Center, Cleveland, Ohio; and 74
6
15 Division of Gastroenterology, McGill University, Montréal, Québec, Canada 75
16 76
17 77
18
19
T his document presents the official recommendations
of the American Gastroenterological Association
(AGA) on the initial management of acute pancreatitis (AP).
recommendations in this guideline. There are 2 overlapping
phases of AP, early and late. The early phase of AP takes
place in the first 2 weeks after disease onset, and the late
78
79
20 The guideline was developed by the AGA’s Clinical Practice phase can last weeks to months thereafter.9 80
21 Guideline Committee and approved by the AGA Governing In this guideline, we address the initial management of 81
22 Board. It is accompanied by a technical review that is a AP within the first 48
72 hours of admission. We focus on 82
23 compilation of the clinical evidence from which these rec- the initial management of AP, as this is the period when 83
24 ommendations were formulated.1 management decisions can alter the course of disease and 84
25 AP is an inflammatory condition of the pancreas that can duration of hospitalization. The management of AP has 85
26 cause local injury, systemic inflammatory response syn- evolved slowly during the preceding 100 years. However, 86
27 drome, and organ failure. Worldwide, AP is a common emerging evidence challenges many of the long-held man- 87
28 gastrointestinal condition that is associated with substantial agement paradigms in AP regarding the benefit of antibi- 88
29 suffering, morbidity, and cost to the health care system. In otics, the timing and mode of nutritional support, and the 89
30 the United States, AP is a leading cause of inpatient care utility and timing of endoscopic retrograde chol- 90
31 among gastrointestinal conditions: >275,000 patients are angiopancreatography (ERCP) and cholecystectomy. There- 91
32 hospitalized for AP annually, at an aggregate cost of >$2.6 fore, we sought to evaluate the sum of the evidence for these 92
33 billion per year.2 The incidence of AP ranges from 5 to 30 and other important questions regarding the management 93
34 cases per 100,000, and there is evidence that the incidence of AP. 94
35 has been rising in recent years.3–5 The overall case fatality Because of the focus on initial treatment of AP, certain 95
36 rate for AP is roughly 5%, and is expectedly higher for more questions pertaining to late complications of AP (eg, man- 96
37 severe disease.6 Patients with AP frequently experience agement of pancreatic fluid collections) are beyond the 97
38 abdominal pain, nausea, and vomiting, and the condition scope of this guideline. Additionally, because this guideline 98
39 negatively impacts quality of life.7 The most common causes focuses on the management of AP, we will not address 99
40 of AP remain gallstones and alcohol, which together diagnostic questions, such as the use of laboratory tests or 100
41 comprise 80% of cases; the remainder of cases are due to radiographic studies to establish the diagnosis of AP. 101
42 less common causes, including drug reactions, pancreatic The guideline was developed utilizing a process outlined 102
43 solid and cystic malignancies, and hypertriglyceridemia.8 elsewhere.11 Briefly, the AGA process for developing clinical 103
44 The diagnosis of AP requires at least 2 of the following practice guidelines incorporates Grading of Recommenda- 104
45 features: characteristic abdominal pain; biochemical evi- tions Assessment, Development and Evaluation (GRADE) 105
46 dence of pancreatitis (ie, amylase or lipase elevated >3 methodology12 and best practices as outlined by the Insti- 106
47 times the upper limit of normal); and/or radiographic evi- tute of Medicine.13 GRADE methodology was utilized to 107
48 dence of pancreatitis on cross-sectional imaging.9 Pre- prepare the background information for the guideline and 108
AGA SECTION

49 sentations of AP occur along a clinical spectrum, and can be the technical review that accompanies it.1 Optimal under- 109
50 categorized as mild, moderately severe, or severe, based on standing of this guideline will be enhanced by reading 110
51 the recent revised Atlanta classification.9 Most cases of AP applicable portions of the technical review. The guideline 111
52 (around 80%)10 are mild, with only interstitial changes of 112
53 the pancreas without local or systemic complications. 113
54 Moderately severe pancreatitis is characterized by transient Abbreviations used in this paper: AGA, American Gastroenterological 114
55 local or systemic complications or transient organ failure Association; AP, acute pancreatitis; CI, confidence interval; ERCP, 115
endoscopic retrograde cholangiopancreatography; GRADE, Grading of
56 (<48 hours), and severe AP is associated with persistent Recommendations Assessment, Development and Evaluation; HES, 116
57 organ failure.9 Necrotizing pancreatitis is characterized by hydroxyethyl starch; OR, odds ratio; RCT, randomized controlled trial. 117
58 the presence of pancreatic and/or peripancreatic necrosis, 118
59 and is typically seen in patients with moderately severe or © 2018 by the AGA Institute
0016-5085/$36.00
119
60 severe AP. Severity of disease factors into several of the https://doi.org/10.1053/j.gastro.2018.01.032 120

PGL 5.5.0 DTD
YGAST61641_proof
3 February 2018
7:14 pm
ce

 2 Crockett et al Gastroenterology Vol. -, No. -

121 Table 1.Quality of Evidence Categories Fluid therapy to prevent hypovolemia and organ hypo- 181
122 perfusion is a long-established cornerstone of the initial 182
Quality of evidence Interpretation management of AP. However, the evidence basis for fluid
123 183
124 High We are very confident that the true effect therapy in AP is relatively weak. In the technical review, a 184
125 lies close to that of the estimate of total of 7 randomized trials were identified pertaining to 185
126 the effect. fluid resuscitation, with 4 primarily addressing the role of 186
127
Moderate We are moderately confident in the effect goal-directed targeted therapy.1 Goal-directed therapy is 187
128
estimate. The true effect is likely to be generally defined as the titration of intravenous fluids to 188
close to the estimate of the effect, but specific clinical and biochemical targets of perfusion (eg,
129 there is a possibility that it is substantially 189
130 heart rate, mean arterial pressure, central venous pressure, 190
different.
131 Low Our confidence in the effect estimate is urine output, blood urea nitrogen concentration, and he- 191
132 limited. The true effect may be matocrit). Use of goal-directed therapy has been shown to 192
133 substantially different from the estimate lower mortality in sepsis,14 a condition with physiologic 193
of the effect. similarities to AP. Compared to non-targeted therapy, goal-
134 Very low We have very little confidence in the effect
194
135 directed therapy did not result in significantly improved 195
estimate. The true effect is likely to be mortality, prevention of pancreatic necrosis, or decrease in
136 substantially different from the estimate 196
137 of effect.
the rate of persistent multiple organ failure. In this context, 197
138 though there was not clear randomized controlled trial 198
139 (RCT)
level evidence of benefit, the panel issued a condi- 199
140 tional recommendation suggesting the use of judicious goal- 200
141 directed fluid therapy vs other methods. However, the panel 201
panel and the authors of the technical review met face to recognized that overly aggressive fluid therapy can be
142 face on July 18, 2017, to discuss the findings from the 202
143 associated with harms in AP, including respiratory compli- 203
technical review. The guideline authors subsequently cations and abdominal compartment syndrome.15,16 The
144 formulated the recommendations. Although the quality of 204
145 overall quality of the evidence was very low due to the 205
the evidence (Table 1) was a key factor in determining the inconsistency among reported outcome measures (espe-
146 strength of the recommendations (Table 2), the panel also 206
147 cially the lack of differentiation between transient and 207
considered the balance between benefit and harm of in- persistent organ failure), the small number of RCTs,
148 terventions, patients’ values and preferences, and resource 208
149 outcome assessment (detection bias), and lack of blinding 209
utilization. The recommendations are summarized in (performance bias). The lack of RCT evidence addressing the
150 Table 3. 210
151 optimal initial rate, volume, and duration of fluid resusci- 211
152 tation in AP rendered the panel unable to make specific 212
Recommendation 1A. In patients with AP, the AGA recommendations in this regard.
153 213
suggests using goal-directed therapy for fluid Regarding the use of Ringer’s lactate vs normal saline as
154 management. Conditional recommendation, very low 214
155 the optimal fluid solution for resuscitation, the panel could 215
quality evidence.
156 Comment: The AGA makes no recommendation whether
not make a recommendation based on the low quality of 216
157 normal saline or Ringer’s lactate is used. evidence. The 2 RCTs specifically addressing this topic used 217
158 surrogate markers of severity and did not focus on 218
159 219
160 220
Table 2.Interpretation of Strength of Recommendation Categories
161 221
162 Strength of Wording in the For the For the 222
163 recommendation guideline patient clinician 223
164 224
165 Strong “The AGA recommends.” Most individuals in this situation would Most individuals should receive the 225
want the recommended course of recommended course of action. Formal
166 226
action and only a small proportion decision aids are not likely to be needed
167 would not. to help individuals make decisions
227
168 consistent with their values and preferences. 228
AGA SECTION

169 Conditional “The AGA suggests.” The majority of individuals in this Different choices will be appropriate for 229
170 situation would want the suggested different patients. Decision aids may be 230
171 course of action, but many would not. useful in helping individuals in making 231
172 decisions consistent with their values and 232
preferences. Clinicians should expect to
173 233
spend more time with patients when
174 working toward a decision. 234
175 No recommendation “The AGA makes no The confidence in the effect estimate is so 235
176 recommendation.” low that any recommendation is 236
177 speculative at this time 237
178 238
179 239
180 240

PGL 5.5.0 DTD
YGAST61641_proof
3 February 2018
7:14 pm
ce

 - 2018 AGA Guideline on Initial Management of AP 3

241 Table 3.Summary of Recommendations of the American Gastroenterological Association Clinical Guidelines for the Initial 301
Management of Acute Pancreatitis
242 302
243 Strength of Quality of 303
244 Recommendation recommendation evidence 304
245 305
246 1A. In patients with AP, the AGA suggests using goal-directed therapy for fluid management. Conditional Very low 306
247 Comment: The AGA makes no recommendation whether normal saline or Ringer’s lactate is used. 307
1B. In patients with AP, the AGA suggests against the use of HES fluids. Conditional Very low
248 2. In patients with predicted severe AP and necrotizing AP, the AGA suggests against the Conditional Low
308
249 use of prophylactic antibiotics. 309
250 3. In patients with acute biliary pancreatitis and no cholangitis, the AGA suggests against the Conditional Low 310
251 routine use of urgent ERCP. 311
252 4. In patients with AP, the AGA recommends early (within 24 h) oral feeding as tolerated, rather Strong Moderate 312
253 than keeping the patient nil per os. 313
5. In patients with AP and inability to feed orally, the AGA recommends enteral rather than Strong Moderate
254 314
parenteral nutrition.
255 6. In patients with predicted severe or necrotizing pancreatitis requiring enteral tube feeding, Conditional Low
315
256 the AGA suggest either NG or NJ route. 316
257 7. In patients with acute biliary pancreatitis, the AGA recommends cholecystectomy during the Strong Moderate 317
258 initial admission rather than after discharge. 318
259 8. In patients with acute alcoholic pancreatitis, the AGA recommends brief alcohol intervention Strong Moderate 319
260 during admission 320
261 321
262 NG, nasogastric; NJ, nasojejunal. 322
263 323
264 important clinical outcomes, such as organ failure, necrosis, and a trend toward reduction in mortality (OR, 0.66; 95% 324
265 or mortality. The panel recognizes that the current intensive CI, 0.42
1.04). However, in a subgroup analysis that 325
266 study of this topic may lead to changing this recommenda- included only recent trials published after 2002, no differ- 326
267 tion in the near future. ences in risks of infected pancreatic and peripancreatic 327
268 necrosis (OR, 0.81; 95% CI, 0.44
1.49) or mortality (OR, 328
269 Recommendation 1B. In patients with AP, the AGA 0.85; 95% CI, 0.52
1.8) were noted. Similarly, there were 329
270 suggests against the use of hydroxyethyl starch no differences in these 2 critical outcomes among higher- 330
271 (HES) fluids. Conditional recommendation, very low quality studies. Given the higher methodologic quality of 331
272 quality evidence.
the recent studies, the guideline panel placed greater 332
273 emphasis on results published after 2002 for this recom- 333
274 mendation. Prophylactic antibiotics had no impact on the 334
The technical review revealed few studies that specif-
275 rates of important outcomes, such as persistent single organ 335
ically addressed the issue of using HES as a resuscitative
276 336
fluid in AP.1 The panel’s conditional recommendation failure, multiple organ failure or multiple organ dysfunction
277 337
against using HES fluids is based on 2 studies examining this of unclear duration, single organ failure of unclear duration,
278 338
issue,17,18 with mortality not improved compared to fluid and hospital length of stay. Though this recommendation
279 statement is specific for patients with severe AP, it should 339
resuscitation without HES. Importantly, multiple organ
280 340
failure was significantly increased in 1 trial with HES fluids be clarified that there is also no role for prophylactic anti-
281 biotics in patients with milder forms of AP. The overall 341
(odds ratio [OR], 3.86; 95% confidence interval [CI],
282 342
1.24
12.04).18 Unfortunately, other important outcomes, quality of evidence was graded as low because of meth-
283 odologic limitations (ie, risk of bias due to lack of blinding of 343
such as development of necrosis and/or persistent organ
284 344
failure were not evaluated in these studies. These findings in participants and study personnel and imprecision).
285 345
AP mirror recent studies in the critical care literature, which
286 Recommendation 3. In patients with acute biliary 346
have not demonstrated a mortality benefit of HES-
287 pancreatitis and no cholangitis, the AGA suggests 347
containing fluids as resuscitative agents.19
288 against the routine use of urgent ERCP. Conditional 348
AGA SECTION

289 Recommendation 2. In patients with predicted

recommendation, low quality evidence. 349
290 severe AP and necrotizing pancreatitis, the AGA 350
291 suggests against the use of prophylactic antibiotics. 351
A total of 8 RCTs addressed the role of urgent ERCP in
292 Conditional recommendation, low quality evidence. 352
the management of patients with acute gallstone pancrea-
293 353
titis.1 Compared to conservative management, urgent ERCP
294 354
295 The technical review,1 which included 10 RCTs had no impact on critical outcomes, such as mortality and
355
addressing the role of prophylactic antibiotics in patients multiple organ failure, and on important outcomes, such as
296 356
297 with predicted severe AP and necrotizing pancreatitis, single organ failure (eg, respiratory or renal), infected 357
298 demonstrated a reduction in the risk of infected pancreatic pancreatic and peripancreatic necrosis, and total rates of 358
299 and peripancreatic necrosis (OR, 0.56; 95% CI, 0.36
0.86) necrotizing pancreatitis. Similar findings were noted in a 359
300 360

PGL 5.5.0 DTD
YGAST61641_proof
3 February 2018
7:14 pm
ce

 4 Crockett et al Gastroenterology Vol. -, No. -

361 subgroup analysis of studies that clearly excluded patients The technical review identified 12 RCTs that compared 421
362 with biliary obstruction. The guideline panel acknowledged the use of parenteral (ie, total parenteral nutrition) vs 422
363 the results of a single study demonstrating a reduction in enteral (oral or enteral tube) feeding in patients with AP. 423
364 hospital length of stay, but the overall body of evidence for There was clear evidence to support the benefit of enteral 424
365 this end point is sparse. The overall quality of evidence was nutrition over total parenteral nutrition with respect to 425
366 graded as low given the inconsistency of results, indirect- reduced risk of infected peripancreatic necrosis (OR, 0.28; 426
367 ness of the evidence, and imprecision of results. The panel 95% CI, 0.15
0.51), single organ failure (OR, 0.25; 95% CI, 427
368 also acknowledged the limitations of published studies 0.10
0.62), and multiple organ failure (OR, 0.41; 95% CI, 428
369 in excluding patients with acute cholangitis (a clear indica- 0.27
0.63). The AGA issued a strong recommendation 429
370 tion for ERCP in patients with or without acute biliary based on the overall moderate quality of available evidence, 430
371 pancreatitis). and the likelihood of increased harm associated with the 431
372 unnecessary use of parenteral nutrition. 432
373 Recommendation 4. In patients with AP, the AGA 433
recommends early (within 24 hours) oral feeding as Recommendation 6. In patients with predicted severe
374 or necrotizing pancreatitis requiring enteral tube
434
375 tolerated rather than keeping the patient nil per os.
feeding, the AGA suggests either nasogastric or 435
Strong recommendation; moderate quality evidence.
376 nasoenteral route. Conditional recommendation, low 436
377 quality evidence. 437
378 Traditional dogma regarding management of AP pre- 438
379 scribed “bowel rest” in an attempt to avoid further stim- 439
380 Three RCTs were identified in the technical review that 440
ulation of the inflamed pancreas. However, current
381 specifically addressed the issue of nasogastric vs nasoen- 441
evidence demonstrates the benefit of the opposite
382 teral (either nasoduodenal or nasojejunal) feeding in AP.1 442
approach, that is, early feeding. Maintaining enteral nutri-
383 The trials did not demonstrate a mortality benefit associ- 443
tion is thought to help protect the gut
mucosal barrier and
384 ated with either modality (OR, 1.01; 95% CI, 0.44
2.30), but 444
reduce bacterial translocation, thereby reducing the risk of
385 there were several methodologic issues that made the evi- 445
infected peripancreatic necrosis and other serious AP
386 dence of low quality, including a small number of RCTs, high 446
outcomes.20
387 risk of performance bias due to participant blinding, and a 447
Combined results of 11 RCTs that addressed the role of
388 high risk of detection bias due to issues with outcome 448
early vs delayed feeding demonstrated no difference in
389 assessment. The studies also did not adequately address the 449
mortality for early vs delayed feeding. There was, however,
390 issue of safety, including aspiration risk, with either of these 450
a 2.5-fold higher risk of interventions for necrosis associ-
391 modalities. The panel recognizes that safety concerns 451
ated with delayed vs early feeding (OR, 2.47; 95% CI,
392 regarding the risk of aspiration may preclude practitioners 452
1.41
4.35), as well as trends observed for higher rates of
393 from using nasogastric tubes in patients with severe AP. 453
infected peripancreatic necrosis (OR, 2.69; 95% CI,
394 0.80
3.60), multiple organ failure (OR, 2.00; 95% CI, 454
Recommendation 7. In patients with acute biliary
395 0.49
8.22), and total necrotizing pancreatitis (OR, 1.84; 455
pancreatitis, the AGA recommends cholecystectomy
396 95% CI, 0.88
3.86) associated with delayed feeding. Based during the initial admission rather than after 456
397 on these studies, the AGA recommends initiation of early discharge. Strong recommendation, moderate 457
398 oral feeding (generally within 24 hours) instead of keeping quality evidence. 458
399 patients NPO. While type of diet was not specifically 459
400 examined in the technical review, success of early feeding 460
401 Cholecystectomy can clearly prevent recurrent episodes 461
has been demonstrated using a variety of diets including
402 of AP after an index case of biliary or gallstone pancrea- 462
low-fat, normal fat, and soft or solid consistency,21 and thus
403 titis.22 However, the appropriate timing of cholecystectomy 463
starting with a clear liquid diet is not required. The panel
404 in patients with biliary or gallstone pancreatitis has been 464
recognized that early feeding is not successful in all AP
405 the subject of vigorous debate. The primary argument in 465
patients due to pain, vomiting, or ileus, and feeding may
406 favor of earlier intervention is that patients with biliary 466
need to be delayed beyond 24 hours in some cases.
407 pancreatitis who are discharged without a cholecystectomy 467
Furthermore, some patients who are intolerant of oral
408 have a significant risk of recurrent biliary events.23 How- 468
feeding may require placement of an enteral tube for
AGA SECTION

409 ever, those who advocate delayed cholecystectomy argue 469

nutritional support (see Recommendation 5). However,
410 that performing surgery at a later time point when the acute 470
routine or empiric orders for nil per os status in patients
411 inflammatory state of AP has subsided may be safer and 471
with AP should generally be avoided in favor of feeding
412 associated with better surgical outcomes. 472
trials. This is a strong recommendation based on the mod-
413 Moderate quality evidence from a single randomized 473
erate quality evidence underpinning the statement.
414 controlled clinical trial24 found that cholecystectomy per- 474
415 formed during the initial admission for patients with sus- 475
Recommendation 5. In patients with AP and inability to pected biliary pancreatitis was associated with substantial
416 feed orally, the AGA recommends enteral rather than 476
417 reductions in a composite outcome of mortality and 477
parenteral nutrition. Strong recommendation,
418 moderate quality evidence. gallstone-related complications (OR, 0.24; 95% CI, 478
419 0.09
0.61), readmission for recurrent pancreatitis (OR, 479
420 480

PGL 5.5.0 DTD
YGAST61641_proof
3 February 2018
7:14 pm
ce

 - 2018 AGA Guideline on Initial Management of AP 5

481 0.25; 95% CI, 0.07
0.90), and pancreaticobiliary compli- high-quality and high-value care for patients with AP. Cur- 541
482 cations (OR, 0.24; 95% CI, 0.09
0.61). Same-admission rent evidence supports the benefit of goal-directed fluid 542
483 cholecystectomy did not differ from delayed cholecystec- resuscitation, early oral feeding, and enteral rather than 543
484 tomy with respect to rates of conversion from laparoscopy parenteral nutrition, in all patients with AP. Our evidence 544
485 to open approach or surgical difficulty. The AGA issued a profiles also support the benefit of same-admission chole- 545
486 strong recommendation due to the quality of available evi- cystectomy for patients with biliary pancreatitis, and brief 546
487 dence and the high likelihood of benefit from early vs alcohol intervention for patients with alcohol-induced 547
488 delayed cholecystectomy in this patient population. pancreatitis. In contrast, current evidence does not sup- 548
489 port a benefit for the routine use of prophylactic antibiotics 549
Recommendation 8. In patients with acute alcoholic in predicted severe AP or routine ERCP in patients with AP
490 pancreatitis, the AGA recommends brief 550
491 without accompanying cholangitis. 551
alcohol intervention during admission. Strong There are several knowledge gaps in the initial manage-
492 recommendation, moderate quality evidence. 552
493 ment of AP that have been identified for which RCTs are 553
494 warranted, as is highlighted in the technical review that ac- 554
495 The technical review identified significant knowledge companies this guideline.1 More evidence is needed to 555
496 gaps in this field with a paucity of RCTs addressing the role deterimine the optimal fluid therapy practice in AP, and to 556
497 of alcohol counseling.1 The panel’s decision to provide a better quantify the benefits and harms of goal-directed 557
498 strong recommendation for a brief alcohol counseling therapy vs other approaches. Current evidence does not 558
499 intervention during admission was driven by the following support a clear benefit of Ringer’s lactate solution compared 559
500 published studies. A single RCT addressed the role of to normal saline for important outcomes, such as organ fail- 560
501 alcohol counseling on recurrent attacks of AP in patients ure, necrosis, or mortality. Future RCTs addressing this topic 561
502 with a first attack of AP with a clear history of alcohol use would provide helpful guidance in this regard. Though risk 562
503 and exclusion of other possible etiologies.25 Patients were stratification of patients with AP is important to ensure 563
504 randomized to either repeated intervention at 6-month in- appropriate level of care, there is a dearth of high-quality 564
505 tervals for 2 years at an outpatient gastrointestinal clinic or evidence measuring the actual clinical impacts of using any 565
506 single intervention at initial hospitalization. There was a particular severity prediction tool. High-quality multicenter 566
507 strong trend toward a reduction for total hospital admission RCTs are required to determine whether prophylactic anti- 567
508 rates with no statistically significant differences for out- biotics have a role in specific groups of patients with pre- 568
509 comes, such as second attack of pancreatitis, definite dicted severe AP and necrotizing pancreatitis. The 569
510 recurrent pancreatitis, or 2 recurrent attacks of pancrea- appropriate timing of ERCP in patients with predicted severe 570
511 titis. The second source of evidence that supports this biliary pancreatitis with persistent biliary obstruction also 571
512 recommendation was a Cochrane review of alcohol reduc- needs to be clarified in future studies. In addition, future 572
513 tion strategies in primary care populations (21 RCTs, n ¼ research should focus on the impact of alcohol and tobacco 573
514 7286), although not specifically addressing patients with cessation interventions on end points, such as recurrent AP, 574
515 AP.26 This study showed that individuals receiving a brief progression to chronic pancreatitis and pancreatic cancer, 575
516 intervention reduced alcohol consumption compared to the quality of life, health care utilization, and mortality. 576
517 control group (mean difference:
41 g/wk; 95% CI,
57 577
518 to
25 g/wk), with substantial heterogeneity in results. 578
519 Extended intervention compared to brief intervention was References 579
520 associated with a nonsignificantly greater reduction in 1. Vege S, DiMagno M, Forsmark CE, et al. Initial medical 580
521 alcohol consumption. Finally, A follow-up meta-analysis treatment of acute pancreatitis: AGA Institute Technical 581
522 addressing the effectiveness of brief interventions in pri- Review. Gastroenterology 2018. 582
523 mary care and differences between efficacy and effective- 2. Peery AF, Crockett SD, Barritt AS, et al. Burden of 583
524 ness trials demonstrated similar results in reduction in gastrointestinal, liver, and pancreatic diseases in the 584
525 alcohol consumption in participants receiving a brief inter- United States. Gastroenterology 2015;149:1731–1741 e3. 585
526 vention and no significant difference in effect sizes for ef- 3. Yadav D, Lowenfels AB. Trends in the epidemiology of 586
527 ficacy and effectiveness trials.27 The overall evidence for the first attack of acute pancreatitis: a systematic review. 587
528 this recommendation was graded down to moderate, given Pancreas 2006;33:323–330. 588
AGA SECTION

529 the indirectness of evidence, risk of bias associated with lack 4. Fagenholz PJ, Castillo CF, Harris NS, et al. Increasing 589
530 of blinding, and imprecision of results. United States hospital admissions for acute pancreatitis, 590
531 1988-2003. Ann Epidemiol 2007;17:491–497. 591
532 5. Krishna SG, Kamboj AK, Hart PA, et al. The changing 592
533
Summary epidemiology of acute pancreatitis hospitalizations: a 593
534 These practice guideline recommendations for the initial decade of trends and the impact of chronic pancreatitis. 594
535 management of AP were developed using the GRADE Pancreas 2017;46:482–488. 595
536 framework and in adherence with the standards for guide- 6. Koutroumpakis E, Slivka A, Furlan A, et al. Management 596
537 line development set forth by the Institute of Medicine for and outcomes of acute pancreatitis patients over the last 597
538 the creation of trustworthy guidelines.11,13 These guidelines decade: a US tertiary-center experience. Pancreatology 598
are intended to reduce practice variation and promote 2017;17:32–40.
539 599
540 600

PGL 5.5.0 DTD
YGAST61641_proof
3 February 2018
7:14 pm
ce

 6 Crockett et al Gastroenterology Vol. -, No. -

7. Pendharkar SA, Salt K, Plank LD, et al. Quality of life after 21. Lankisch PG, Apte M, Banks PA. Acute pancreatitis.
601 661
acute pancreatitis: a systematic review and meta- Lancet 2015;386:85–96.
602 662
analysis. Pancreas 2014;43:1194–1200. 22. Yadav D, O’Connell M, Papachristou GI. Natural history
603 663
8. Gullo L, Migliori M, Olah A, et al. Acute pancreatitis in five following the first attack of acute pancreatitis. Am J
604 664
European countries: etiology and mortality. Pancreas Gastroenterol 2012;107:1096–1103.
605 665
2002;24:223–227. 23. van Baal MC, Besselink MG, Bakker OJ, et al. Timing of
606 666
9. Banks PA, Bollen TL, Dervenis C, et al. Classification of cholecystectomy after mild biliary pancreatitis: a sys-
607 667
acute pancreatitis—2012: revision of the Atlanta classi- tematic review. Ann Surg 2012;255:860–866.
608 668
fication and definitions by international consensus. Gut 24. da Costa DW, Bouwense SA, Schepers NJ, et al. Same-
609 2013;62:102–111. 669
admission versus interval cholecystectomy for mild
610 670
10. van Dijk SM, Hallensleben NDL, van Santvoort HC, et al. gallstone pancreatitis (PONCHO): a multicentre rando-
611 Acute pancreatitis: recent advances through randomised mised controlled trial. Lancet 2015;386:1261–1268. 671
612 trials. Gut 2017;66:2024–2032. 672
25. Nordback I, Pelli H, Lappalainen-Lehto R, et al. The
613 11. American Gastroenterological Association Institute Clin- recurrence of acute alcohol-associated pancreatitis can 673
614 ical Practice Guideline Development Process. http://www. be reduced: a randomized controlled trial. Gastroenter- 674
615 gastro.org/guidelines/guidelines-policies. Accessed July ology 2009;136:848–855. 675
616 31, 2017. 26. Kaner EF, Beyer F, Dickinson HO, et al. Effectiveness of 676
617 12. Sultan S, Falck-Ytter Y, Inadomi JM. The AGA institute brief alcohol interventions in primary care populations. 677
618 process for developing clinical practice guidelines part Cochrane Database Syst Rev 2007:CD004148. 678
619 one: grading the evidence. Clin Gastroenterol Hepatol 27. Kaner EF, Dickinson HO, Beyer F, et al. The effectiveness 679
620 2013;11:329–332. of brief alcohol interventions in primary care settings: a 680
621 13. Institute of Medicine. Clinical Practice Guidelines We Can systematic review. Drug Alcohol Rev 2009;28:301–323. 681
622 Trust. Washington, DC: The National Academies Press, 682
623 2011. 683
624 14. Rivers E, Nguyen B, Havstad S, et al. Early goal-directed 684
Reprint requests
625 therapy in the treatment of severe sepsis and septic Address requests for reprints to: Seth D. Crockett, MD, MPH, Division of 685
626 shock. N Engl J Med 2001;345:1368–1377. Gastroenterology and Hepatology, University of North Carolina, CB 7080, 686
130 Mason Farm Road, Chapel Hill, North Carolina 27599. e-mail:
627 15. de-Madaria E, Soler-Sala G, Sanchez-Paya J, et al. sethc@med.unc.edu; fax: (919) 966-8929. Q1 687
628 Influence of fluid therapy on the prognosis of acute 688
Acknowledgments
629 pancreatitis: a prospective cohort study. Am J Gastro- American Gastroenterological Association Institute Clinical Guidelines
689
630 enterol 2011;106:1843–1850. Committee: Seth Crockett, University of North Carolina at Chapel Hill, 690
631 16. De Waele JJ, Leppaniemi AK. Intra-abdominal hyper- Chapel Hill, NC; Yngve Falck-Ytter, VA and Case Medical Center, Cleveland,
OH; Joseph Feuerstein, Beth Israel Deaconess Medical Center, Brookline,
691
632 tension in acute pancreatitis. World J Surg 2009; MA; Steven Flamm, Northwestern University, Chicago, IL; Ziad Gellad, Duke 692
633 33:1128–1133. University Medical Center, Durham, NC; Lauren Gerson, California Pacific 693
Medical Center, San Francisco, CA; Samir Gupta, University of California,
634 17. Du XJ, Hu WM, Xia Q, et al. Hydroxyethyl starch Veterans Affairs Healthcare System, San Diego, CA; Ikuo Hirano, 694
635 resuscitation reduces the risk of intra-abdominal hyper- Northwestern University School of Medicine, Chicago, IL; John Inadomi, 695
University of Washington, Seattle, WA; Geoffrey C. Nguyen, Mount Sinai
636 tension in severe acute pancreatitis. Pancreas 2011; Hospital, University of Toronto, Toronto, ON; Joel H. Rubenstein, Veterans 696
637 40:1220–1225. Affairs Center for Clinical Management Research, Ann Arbor, MI and Division 697
638 18. Zhao G, Zhang JG, Wu HS, et al. Effects of different of Gastroenterology, University of Michigan Medical School, Ann Arbor,
Michigan; Siddharth Singh, University of California San Diego, San Diego,
698
639 resuscitation fluid on severe acute pancreatitis. World J CA; Walter E. Smalley, Vanderbilt University School of Medicine, Nashville, 699
640 Gastroenterol 2013;19:2044–2052. TN; Neil Stollman, University of California San Francisco, Northern California 700
Gastroenterology Consultants, San Francisco, CA; Sarah Street, Stanford
641 19. Myburgh JA, Finfer S, Bellomo R, et al. Hydroxyethyl University, Orinda, CA; Shahnaz Sultan, Minneapolis VA Health Care System, 701
642 starch or saline for fluid resuscitation in intensive care. University of Minnesota, Minneapolis, MN; Santhi S. Vege, Pancreas Group, 702
N Engl J Med 2012;367:1901–1911. Division of Gastroenterology and Hepatology, Mayo Clinic, Rochester, MN;
643 Sachin B. Wani, University of Colorado Anschutz Medical Campus, Aurora, 703
644 20. Windsor AC, Kanwar S, Li AG, et al. Compared with CO; David Weinberg, Department of Medicine, Fox Chase Cancer Center, 704
parenteral nutrition, enteral feeding attenuates the acute Philadelphia, PA.
645 705
646 phase response and improves disease severity in acute Conflicts of interest 706
647 pancreatitis. Gut 1998;42:431–435. The authors disclose no conflicts. Q2
707
648 708
AGA SECTION

649 709
650 710
651 711
652 712
653 713
654 714
655 715
656 716
657 717
658 718
659 719
660 720

PGL 5.5.0 DTD
YGAST61641_proof
3 February 2018
7:14 pm
ce