Pharmacological interventions for acute pancreatitis

Elisabetta Moggia; Rahul Koti; Ajay P Belgaumkar; Federico Fazio; Stephen P Pereira; Brian R Davidson; Kurinchi Selvan Gurusamy

doi:10.1002/14651858.CD011384.pub2

Intervenciones farmacológicas para la pancreatitis aguda

Authors' declarations of interest

Version published: 21 April 2017 Version history

https://doi.org/10.1002/14651858.CD011384.pub2

Collapse all Expand all

Resumen

available in

Antecedentes

En los pacientes con pancreatitis aguda, no está claro cuál debería ser la función del tratamiento médico sumado a la atención de apoyo, como el equilibrio hidroelectrolítico y el apoyo a los órganos en pacientes con insuficiencia orgánica.

Objetivos

Evaluar los efectos de diferentes intervenciones farmacológicas en pacientes con pancreatitis aguda.

Métodos de búsqueda

Se hicieron búsquedas en el Registro Cochrane Central de Ensayos Controlados (Cochrane Central Register of Controlled Trials) (CENTRAL, 2016, número 9), MEDLINE, Embase, Science Citation Index Expanded, y en registros de ensayos hasta octubre 2016 para identificar ensayos controlados aleatorios (ECA). También se buscaron las referencias de los ensayos incluidos para identificar ensayos adicionales.

Criterios de selección

Se consideró la inclusión en la revisión de ECA realizados en pacientes con pancreatitis aguda solamente, de manera independiente de la etiología, la gravedad, la presencia de infección, el idioma, el cegamiento o el estado de publicación.

Obtención y análisis de los datos

Dos autores de la revisión de forma independiente identificaron los ensayos y extrajeron los datos. No se realizó el metanálisis en red programado por la falta de información sobre los modificadores del efecto potencial y las diferencias en el tipo de participantes incluidos en las diferentes comparaciones, cuando hubo información disponible. Se calculó el odds ratio (OR) con intervalos de confianza (IC) del 95% para los resultados binarios y los cocientes de tasas con IC del 95% para los resultados de los recuentos mediante un modelo de efectos fijos y un modelo de efectos aleatorios.

Resultados principales

En esta revisión se incluyeron 84 ECA con 8234 participantes. Seis ensayos (N = 658) no informaron ninguno de los resultados de interés para esta revisión. Los 78 ensayos restantes excluyeron a 210 participantes después de la asignación al azar. En consecuencia, un total de 7366 participantes en 78 ensayos contribuyeron a uno o más resultados para esta revisión. Los tratamientos evaluados en estos 78 ensayos fueron: antibióticos, antioxidantes, aprotinina, atropina, calcitonina, cimetidina, EDTA (ácido etilendiaminotetraacético), gabexato, glucagón, iniprol, lexipafant, AINE (fármacos antiinflamatorios no esteroideos), octreotida, oxifenonio, probióticos, proteína C activada, somatostatina, somatostatina más omeprazol, somatostatina más ulinastatina, timosina, ulinastatina y control inactivo. Además de la comparación de antibióticos versus control, que incluyó a gran parte de los participantes con pancreatitis necrosante, en las comparaciones restantes sólo hubo una proporción pequeña de pacientes con este trastorno. La mayoría de los ensayos incluyeron sólo a pacientes con pancreatitis aguda grave o una combinación de pacientes con pancreatitis aguda leve y pancreatitis aguda grave (75 ensayos). En términos generales, el riesgo de sesgo en los ensayos fue poco claro o alto para todos los ensayos excepto uno.

Fuente de financiación: siete ensayos no fueron financiados o fueron financiados por organismos sin intereses personales en los resultados. Las compañías farmacéuticas financiaron parcial o totalmente 21 ensayos. La fuente de financiación no estaba disponible en los ensayos restantes.

Debido a que la mortalidad a corto plazo se consideró el resultado más importante, en el resumen sólo se presentaron estos resultados de forma detallada. Sesenta y siete estudios con 6638 participantes informaron la mortalidad a corto plazo. No hubo pruebas de diferencias en la mortalidad a corto plazo en ninguna de las comparaciones (evidencia de muy baja calidad). Con respecto a otros resultados primarios, los eventos adversos graves (número) fueron inferiores que en el control en los participantes que recibieron lexipafant (cociente de tasas 0,67; IC del 95%: 0,46 a 0,96; N = 290; un estudio; evidencia de muy baja calidad), octreotida (cociente de tasas 0,74; IC del 95%: 0,60 a 0,89; N = 770; cinco estudios; evidencia de muy baja calidad), somatostatina más omeprazol (cociente de tasas 0,36; IC del 95%: 0,19 a 0,70; N = 140; un estudio; evidencia de baja calidad) y somatostatina más ulinastatina (cociente de tasas 0,30; IC del 95%: 0,15 a 0,60; N = 122; un estudio; evidencia de baja calidad). La proporción de pacientes con insuficiencia orgánica fue inferior en el grupo de octreotida que en el control (OR 0,51; IC del 95%: 0,27 a 0,97; N = 430; tres estudios; evidencia de muy baja calidad). La proporción de pacientes con sepsis fue inferior en el grupo de lexipafant que en el control (OR 0,26; IC del 95%: 0,08 a 0,83; N = 290; un estudio; evidencia de muy baja calidad). No hubo evidencia de diferencias en ninguna de las comparaciones restantes en estos resultados ni para ninguno de los resultados primarios restantes (la proporción de pacientes con al menos un evento adverso grave y la presencia de necrosis pancreática infectada). Ninguno de los ensayos informó la calidad de vida relacionada con la salud.

Conclusiones de los autores

La evidencia de muy baja calidad indica que ninguno de los tratamientos farmacológicos estudió la disminución en la mortalidad a corto plazo en los pacientes con pancreatitis aguda. Sin embargo, los intervalos de confianza fueron amplios y compatibles con un aumento o una disminución en la mortalidad a corto plazo debido a las intervenciones. No se encontraron beneficios clínicos consistentes con ninguna intervención. Dadas las limitaciones en los sistemas de puntuación de pronóstico y debido a que el daño a los órganos puede ocurrir en la pancreatitis aguda antes de que se manifieste clínicamente, los ensayos futuros deben considerar la posibilidad de incluir todos los casos de pancreatitis pero aportar al estudio el poder estadístico suficiente para medir las diferencias en el subgrupo de pacientes con pancreatitis aguda grave. Puede ser difícil que los estudios basados en la mortalidad tengan un poder estadístico suficiente. Los ensayos futuros en pacientes con pancreatitis aguda deben considerar otros resultados, como las complicaciones o la calidad de vida relacionada con la salud como resultados primarios. Estos ensayos deben incluir la calidad de vida relacionada con la salud, los costos y el retorno al trabajo como resultados y deben realizar el seguimiento de los pacientes durante al menos tres meses (preferentemente durante al menos un año).

PICOs

Population

Intervention

Comparison

Outcome

The PICO model is widely used and taught in evidence-based health care as a strategy for formulating questions and search strategies and for characterizing clinical studies or meta-analyses. PICO stands for four different potential components of a clinical question: Patient, Population or Problem; Intervention; Comparison; Outcome.

See more on using PICO in the Cochrane Handbook.

Resumen en términos sencillos

available in

Tratamiento médico para los pacientes con pancreatitis aguda (inflamación aguda del páncreas)

Antecedentes

El páncreas es un órgano ubicado en el abdomen que secreta varias enzimas digestivas (sustancias que permiten y aceleran las reacciones químicas en el cuerpo) en el sistema ductal pancreático, y el contenido luego se vacía en el intestino delgado. También contiene los islotes de Langerhans, que secretan varias hormonas incluida la insulina (que ayuda a regular el azúcar sanguíneo). La pancreatitis aguda es una enfermedad potencialmente mortal caracterizada por la inflamación súbita del páncreas, que puede dar lugar a la insuficiencia de otros órganos, como el pulmón y el riñón. Hay mucha investigación sobre diferentes tratamientos médicos para la pancreatitis aguda, aunque no están claros los beneficios de cada tratamiento ni si algún tratamiento médico es beneficioso además del tratamiento de apoyo. Esta atención incluye hidratación corporal y tratamiento con cuidados intensivos en los pacientes con insuficiencia orgánica (para apoyar los órganos deteriorados). Se intentó resolver este problema mediante la búsqueda de los estudios existentes sobre el tema. Se incluyeron todos los ensayos controlados aleatorios (estudios clínicos en que los pacientes son asignados al azar a uno de dos o más grupos de tratamiento) cuyos resultados se presentaron hasta el 7 de octubre de 2016.

Características de los estudios

En esta revisión, se incluyeron 84 ECA con 8234 participantes. Seis ensayos (658 participantes) no informaron ninguno de los resultados de interés para esta revisión. En los 78 ensayos restantes, 210 participantes fueron excluidos después de la asignación al azar. Así, un total de 7366 participantes en 78 ensayos contribuyeron a uno o más resultados para esta revisión. Además de la comparación de si deben usarse antibióticos, las otras comparaciones incluyeron sólo un porcentaje pequeño de pacientes con necrosis pancreática (una forma sumamente grave de pancreatitis, que resulta en la destrucción pancreática). La mayoría de los ensayos incluyeron sólo la forma grave de la pancreatitis aguda o incluyeron formas tanto leves como graves de pancreatitis.

Fuente de financiación: siete ensayos no fueron financiados o fueron financiados por organismos sin intereses personales en los resultados. Veintiún ensayos fueron financiados parcial o totalmente por las compañías farmacéuticas. La fuente de financiación no estaba disponible en los ensayos restantes.

Calidad de la evidencia

La calidad general de la evidencia fue baja para todas las medidas debido a que el riesgo de sesgo en los ensayos fue poco claro o alto (un error sistemático o desviación de la verdad que afecta los resultados, favoreciendo un tratamiento sobre otro) y a que se trataba de ensayos pequeños. Como resultado, se necesitan más estudios sobre este tema.

Resultados clave

Sesenta y siete estudios con 6638 participantes informaron las muertes a corto plazo. En términos generales, se registró un promedio de un 12% de defunciones entre los pacientes que recibieron sólo atención de apoyo. No hubo evidencia de que ninguno de los tratamientos redujera las muertes a corto plazo. Hubo evidencia de que diversos tratamientos podrían ser beneficiosos en varios resultados; sin embargo, éstos no fueron consistentes, y no es posible establecer conclusiones sobre si alguno de los tratamientos puede ser beneficioso. Ninguno de los ensayos informó la calidad de vida relacionada con la salud.

En conclusión, sobre la base de las pruebas de muy baja calidad, no existe evidencia de que algún tratamiento farmacológico sumado a la atención de apoyo reduzca las muertes a corto plazo. Los ensayos futuros en los pacientes con pancreatitis aguda deben incluir la calidad de vida relacionada con la salud, los costos y el retorno al trabajo como resultados y deben realizar el seguimiento de los pacientes durante al menos tres meses (preferentemente durante al menos un año).

Authors' conclusions

Implications for practice

Very low‐quality evidence suggests that no pharmacological treatment leads to a decrease in short‐term mortality in people with acute pancreatitis. However, the confidence intervals were wide and consistent with an increase or decrease in short‐term mortality. We did not find consistent clinical benefits with any intervention.

Implications for research

Because of the limitations in the prognostic scoring systems and because damage to organs may occur in acute pancreatitis before they are clinically manifest, future trials should consider including pancreatitis of all severity but power the study to measure the differences in the subgroup of people with severe acute pancreatitis. It may be difficult to power the studies based on mortality. Future trials in patients with acute pancreatitis should consider other outcomes such as complications or health‐related quality of life as primary outcomes. Such trials should include health‐related quality of life, costs, and return to work as outcomes and should follow patients for at least three months (preferably for at least one year).

Summary of findings

Open in table viewer

Summary of findings for the main comparison. Summary of findings (mortality)

Pharmacological interventions for treatment of acute severe pancreatitis (mortality)
Patient or population: people with acute pancreatitis Settings: secondary or tertiary setting Intervention: various treatments Control: inactive control
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect (95% CI)	No of participants (studies)	Quality of the evidence (GRADE)
	Assumed risk Inactive control	Corresponding risk Various treatments
Short‐term mortality Follow‐up: up to 3 months	Antibiotics		OR 0.81 (0.57 to 1.15)	1058 (17 studies)	⊕⊝⊝⊝ Very low^a,b,c
	120 per 1000	99 per 1000 (72 to 135)
	Antioxidants		OR 2.01 (0.53 to 7.56)	163 (4 studies)	⊕⊝⊝⊝ Very low^a,b,c
	120 per 1000	215 per 1000 (68 to 508)
	Aprotinin		OR 0.68 (0.40 to 1.14)	651 (7 studies)	⊕⊝⊝⊝ Very low^a,b,c
	120 per 1000	85 per 1000 (52 to 135)
	Calcitonin		OR 0.55 (0.15 to 2.00)	125 (2 studies)	⊕⊝⊝⊝ Very low^1,²^,3
	120 per 1000	69 per 1000 (20 to 214)
	Cimetidine		OR 1.00 (0.06 to 17.18)	40 (1 study)	⊕⊝⊝⊝ Very low^a,b,c
	120 per 1000	120 per 1000 (8 to 701)
	EDTA		OR 0.94 (0.12 to 7.08)	64 (1 study)	⊕⊝⊝⊝ Very low^1,²^,3
	120 per 1000	113 per 1000 (17 to 491)
	Gabexate		OR 0.79 (0.48 to 1.30)	576 (5 studies)	⊕⊝⊝⊝ Very low^a,b,c
	120 per 1000	98 per 1000 (62 to 151)
	Glucagon		OR 0.97 (0.51 to 1.87)	409 (5 studies)	⊕⊝⊝⊝ Very low^1,²^,3
	120 per 1000	117 per 1000 (65 to 203)
	Iniprol		OR 0.14 (0.01 to 1.67)	24 (1 study)	⊕⊝⊝⊝ Very low^a,b,c
	120 per 1000	19 per 1000 (2 to 185)
	Lexipafant		OR 0.55 (0.30 to 1.01)	423 (3 studies)	⊕⊝⊝⊝ Very low^1,²^,3
	120 per 1000	70 per 1000 (40 to 121)
	Octreotide		OR 0.76 (0.47 to 1.23)	927 (6 studies)	⊕⊝⊝⊝ Very low^a,b,c
	120 per 1000	94 per 1000 (60 to 143)
	Probiotics		OR 1.70 (0.87 to 3.30)	358 (2 studies)	⊕⊝⊝⊝ Very low^a,b,c,d
	120 per 1000	188 per 1000 (106 to 310)
	Activated protein C		OR 8.56 (0.41 to 180.52)	32 (1 study)	⊕⊝⊝⊝ Very low^a,b,c
	120 per 1000	539 per 1000 (52 to 961)
	Somatostatin		OR 0.57 (0.29 to 1.10)	493 (6 studies)	⊕⊝⊝⊝ Very low^a,b,c
	120 per 1000	72 per 1000 (39 to 130)
	Somatostatin plus omeprazole		OR 0.23 (0.05 to 1.11)	140 (1 study)	⊕⊝⊝⊝ Very low^a,b,c
	120 per 1000	30 per 1000 (6 to 132)
	Somatostatin plus ulinastatin		OR 0.43 (0.15 to 1.23)	122 (1 study)	⊕⊝⊝⊝ Very low^a,b,c
	120 per 1000	55 per 1000 (20 to 144)
	Thymosin		Not estimable	24 (1 study)	⊕⊝⊝⊝ Very low^a,b,c
	120 per 1000	not estimable
	Ulinastatin		OR 0.45 (0.12 to 1.72)	132 (2 studies)	⊕⊝⊝⊝ Very low^a,b,c
	120 per 1000	58 per 1000 (16 to 190)
Long‐term mortality Follow‐up: 1 year	None of the trials with inactive treatment in the control group reported long‐term mortality.
The basis for the assumed risk* is the average control group proportion across all comparisons. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence intervals; OR: odds ratio; EDTA: ethylenediaminetetraacetic acid.
GRADE Working Group grades of evidence High quality: further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: we are very uncertain about the estimate.
^aRisk of bias: downgraded by one level. ^bImprecision: downgraded one level for wide confidence intervals. ^cImprecision: downgraded one level for small sample size. ^dHeterogeneity: downgraded one level for lack of overlap of confidence intervals and high I².

Open in table viewer

Summary of findings 2. Summary of findings (other primary outcomes)

Pharmacological interventions for treatment of acute severe pancreatitis (other outcomes)
Patient or population: people with acute pancreatitis Settings: secondary or tertiary setting Intervention: various treatments Control: inactive control
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect (95% CI)	No of participants (studies)	Quality of the evidence (GRADE)
	Assumed risk	Corresponding risk
	Inactive control	Various treatments
Serious adverse events (proportion) Follow‐up: up to 3 months	Antibiotics		OR 0.65 (0.37 to 1.15)	304 (5 studies)	⊕⊝⊝⊝ Very low^a,b,c
	147 per 1000	101 per 1000 (60 to 166)
	Antioxidants		OR 1.98 (0.48 to 8.13)	82 (2 studies)	⊕⊝⊝⊝ Very low^a,b,c
	147 per 1000	255 per 1000 (77 to 584)
	EDTA		OR 0.52 (0.11 to 2.39)	64 (1 study)	⊕⊝⊝⊝ Very low^a,b,c
	147 per 1000	83 per 1000 (19 to 292)
	Gabexate		OR 1.31 (0.31 to 5.60)	201 (2 studies)	⊕⊝⊝⊝ Very low^a,b,c
	147 per 1000	185 per 1000 (51 to 492)
	Glucagon		OR 0.29 (0.01 to 7.46)	127 (2 studies)	⊕⊝⊝⊝ Very low^a,b,c
	147 per 1000	48 per 1000 (2 to 563)
	Octreotide		OR 1.73 (0.61 to 4.93)	58 (1 study)	⊕⊝⊝⊝ Very low^a,b,c,d
	147 per 1000	230 per 1000 (95 to 460)
	Somatostatin		OR 1.07 (0.35 to 3.27)	111 (2 studies)	⊕⊝⊝⊝ Very low^a,b,c,d
	147 per 1000	156 per 1000 (57 to 361)
Serious adverse events (number) Follow‐up: up to 3 months	Antibiotics		Rate ratio0.86 (0.68 to 1.07)	716 (12 studies)	⊕⊝⊝⊝ Very low^a,b,c
	437 per 1000	374 per 1000 (298 to 469)
	Antioxidants		Rate ratio0.22 (0.02 to 2.21)	71 (2 studies)	⊕⊝⊝⊝ Very low^a,b,c
	437 per 1000	94 per 1000 (9 to 967)
	Aprotinin		Rate ratio0.79 (0.49 to 1.29)	264 (3 studies)	⊕⊝⊝⊝ Very low^a,b,c
	437 per 1000	345 per 1000 (212 to 562)
	Cimetidine		Rate ratio1.00 (0.20 to 4.95)	60 (1 study)	⊕⊝⊝⊝ Very low^a,b,c
	437 per 1000	437 per 1000 (88 to 2165)
	EDTA		Rate ratio0.94 (0.19 to 4.65)	64 (1 study)	⊕⊝⊝⊝ Very low^a,b,c
	437 per 1000	411 per 1000 (83 to 2034)
	Gabexate		Rate ratio0.86 (0.64 to 1.15)	375 (3 studies)	⊕⊝⊝⊝ Very low^a,b,c
	437 per 1000	375 per 1000 (279 to 503)
	Glucagon		Rate ratio1.00 (0.02 to 50.40)	68 (1 study)	⊕⊝⊝⊝ Very low^a,b,c
	437 per 1000	437 per 1000 (9 to 22027)
	Lexipafant		rate ratio0.67 (0.46 to 0.96)	290 (1 study)	⊕⊝⊝⊝ Very low^a,b,c
	437 per 1000	292 per 1000 (203 to 420)
	Octreotide		Rate ratio0.74 (0.60 to 0.89)	770 (5 studies)	⊕⊝⊝⊝ Very low^a,b,c
	437 per 1000	321 per 1000 (264 to 391)
	Probiotics		Rate ratio0.94 (0.65 to 1.36)	397 (3 studies)	⊕⊝⊝⊝ Very low^a,b,c,d
	437 per 1000	412 per 1000 (286 to 595)
	Somatostatin		Rate ratio1.03 (0.66 to 1.59)	257 (3 studies)	⊕⊝⊝⊝ Very low^a,b,c
	437 per 1000	449 per 1000 (290 to 695)
	Somatostatin plus omeprazole		Rate ratio0.36 (0.19 to 0.70)	140 (1 study)	⊕⊕⊝⊝ Low^a,b
	437 per 1000	159 per 1000 (82 to 308)
	Somatostatin plus ulinastatin		Rate ratio0.30 (0.15 to 0.60)	122 (1 study)	⊕⊕⊝⊝ Low^a,b
	437 per 1000	133 per 1000 (68 to 262)
Organ failure Follow‐up: up to 3 months	Antibiotics		OR 0.78 (0.44 to 1.38)	258 (5 studies)	⊕⊝⊝⊝ Very low^a,b,c
	289 per 1000	241 per 1000 (152 to 360)
	Antioxidants		OR 0.92 (0.39 to 2.12)	163 (4 studies)	⊕⊝⊝⊝ Very low^a,b,c
	289 per 1000	271 per 1000 (138 to 463)
	Gabexate		OR 0.32 (0.01 to 8.25)	50 (1 study)	⊕⊝⊝⊝ Very low^a,b,c
	289 per 1000	115 per 1000 (5 to 770)
	Lexipafant		OR 0.68 (0.36 to 1.27)	340 (2 studies)	⊕⊝⊝⊝ Very low^a,b,c
	289 per 1000	216 per 1000 (128 to 341)
	Octreotide		OR 0.51 (0.27 to 0.97)	430 (3 studies)	⊕⊝⊝⊝ Very low^a,b,c,d
	289 per 1000	173 per 1000 (99 to 284)
	Probiotics		OR 0.80 (0.26 to 2.47)	358 (2 studies)	⊕⊝⊝⊝ Very low^a,b,c,d
	289 per 1000	246 per 1000 (95 to 501)
	Ulinastatin		OR 0.27 (0.01 to 6.67)	129 (2 studies)	⊕⊝⊝⊝ Very low^a,b,c,d
	289 per 1000	100 per 1000 (5 to 731)
Infected pancreatic necrosis Follow‐up: up to 3 months	Antibiotics		OR 0.82 (0.53 to 1.25)	714 (11 studies)	⊕⊝⊝⊝ Very low^a,b,c
	140 per 1000	118 per 1000 (80 to 169)
	Octreotide		OR 0.52 (0.04 to 6.06)	58 (1 study)	⊕⊝⊝⊝ Very low^a,b,c
	140 per 1000	78 per 1000 (7 to 497)
	Probiotics		OR 1.10 (0.62 to 1.96)	397 (3 studies)	⊕⊝⊝⊝ Very low^a,b,c
	140 per 1000	152 per 1000 (92 to 243)
Sepsis Follow‐up: up to 3 months	Antibiotics		OR 0.42 (0.11 to 1.60)	60 (1 study)	⊕⊝⊝⊝ Very low^a,b,c
	122 per 1000	56 per 1000 (15 to 182)
	Aprotinin		OR 1.84 (0.49 to 6.96)	103 (2 studies)	⊕⊝⊝⊝ Very low^a,b,c
	122 per 1000	204 per 1000 (63 to 492)
	Gabexate		OR 1.10 (0.55 to 2.19)	373 (3 studies)	⊕⊝⊝⊝ Very low^a,b,c
	122 per 1000	133 per 1000 (71 to 233)
	Lexipafant		OR 0.26 (0.08 to 0.83)	290 (1 study)	⊕⊝⊝⊝ Very low^a,b,c
	122 per 1000	35 per 1000 (12 to 103)
	Octreotide		OR 0.40 (0.05 to 3.53)	340 (2 studies)	⊕⊝⊝⊝ Very low^a,b,c,d
	122 per 1000	53 per 1000 (6 to 329)
	Probiotics		OR 0.36 (0.10 to 1.36)	62 (1 study)	⊕⊝⊝⊝ Very low^a,b,c
	122 per 1000	48 per 1000 (13 to 159)
Health‐related quality of life	None of the trials reported this outcome.
The basis for the assumed risk* is the average control group proportion across all comparisons. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence intervals; OR = odds ratio; EDTA = ethylenediaminetetraacetic acid.
GRADE Working Group grades of evidence High quality: further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: we are very uncertain about the estimate.
^aRisk of bias: downgraded by one level. ^bImprecision: downgraded one level for wide confidence intervals. ^cImprecision: downgraded one level for small sample size. ^dHeterogeneity: downgraded one level for lack of overlap of confidence intervals and high I².

Background

Description of the condition

The pancreas is an abdominal organ that secretes several digestive enzymes into the pancreatic ductal system before it empties into the small bowel. The pancreas also lodges the Islets of Langerhans, which secrete several hormones including insulin (NCBI 2014). Acute pancreatitis is a sudden inflammatory process in the pancreas, with variable involvement of nearby organs or other organ systems (Bradley 1993). The annual incidence of acute pancreatitis ranges from 5 to 30 per 100,000 population (Roberts 2013; Yadav 2006). There has been an increase in the incidence of acute pancreatitis in the last 10 to 20 years in the UK and USA (Roberts 2013; Yang 2008). Acute pancreatitis is the commonest gastrointestinal (digestive tract) cause of hospital admission in the USA (Peery 2012), and gallstones and alcohol are the two main causes. Approximately, 50% to 70% of acute pancreatitis is caused by gallstones (Roberts 2013; Yadav 2006); these slip into the common bile duct and obstruct the ampulla of Vater (a common channel formed by the union of common bile duct and pancreatic duct), resulting in obstruction to the flow of pancreatic enzymes and leading to activation of trypsinogen within the pancreas and acute pancreatitis (Sah 2013).

Advanced age, male sex, and lower socioeconomic class are associated with higher incidence of acute pancreatitis (Roberts 2013).

Clinicians generally diagnose acute pancreatitis when at least two of the following three features are present (Banks 2013).

Acute onset of a persistent, severe, epigastric pain, often radiating to the back.
Serum lipase activity (or amylase activity) at least three times greater than the upper limit of normal.
Characteristic findings of acute pancreatitis on contrast‐enhanced computed tomography (CECT) and less commonly magnetic resonance imaging (MRI) or transabdominal ultrasonography.

Depending upon the type of inflammation, acute pancreatitis can be classified into interstitial oedematous pancreatitis (diffuse (widespread) or occasionally localised enlargement of the pancreas due to inflammatory oedema as seen on CECT) or necrotising pancreatitis (necrosis involving either the pancreas, peripancreatic tissues, or both) (Banks 2013). Approximately 90% to 95% of people with acute pancreatitis have interstitial oedematous pancreatitis, while the remainder have necrotising pancreatitis (Banks 2013). Necrotising pancreatitis may be sterile or infected (Banks 2013). Various theories exist as to how pancreatic and peripancreatic tissues get infected. These include spread from blood circulation, lymphatics, bile, and the small bowel (duodenum) through the pancreatic duct, as well as movement (translocation) through the large bowel wall (Schmid 1999).

Local complications of acute pancreatitis include acute peripancreatic fluid collection, pancreatic pseudocyst, acute necrotic collection, and walled‐off necrosis (Banks 2013). The systemic complications of acute pancreatitis include worsening of pre‐existing illnesses such as heart or chronic lung disease (Banks 2013). The mortality rates following an attack of acute pancreatitis are between 6% and 20% (Roberts 2013; Yadav 2006), according to severity. Acute pancreatitis can be classified as mild, moderate, or severe, depending on the presence of local or systemic complications, transient organ failure involving one of more of lungs, kidneys, and cardiovascular system (heart and blood vessels) lasting up to 48 hours, or persistent failure of these organs lasting beyond 48 hours. Mild pancreatitis has the best prognosis, and there are no local or systemic complications or organ failure. In moderately severe acute pancreatitis, there may be local or systemic complications or transient organ failure. Severe acute pancreatitis carries the worst prognosis in terms of mortality, and there is persistent organ failure (Banks 2013).

The clinical manifestation of acute pancreatitis is believed to be caused by activation of inflammatory pathways either directly by the pathologic insult or indirectly by activation of trypsinogen (an enzyme that digests protein or a protease), resulting in formation of trypsin, a protease that can break down the pancreas (Sah 2013). This activation of inflammatory pathways manifests clinically as systemic inflammatory response syndrome (SIRS) (Banks 2013; Sah 2013; Tenner 2013). Systemic inflammatory response syndrome is characterised by two or more of the following criteria (Bone 1992).

Temperature of less than 36°C or more than 38°C.
Heart rate less than 90 beats/minute.
Respiratory rate more than 20/min or PCO₂ less than 32 mm Hg.
White blood cell count more than 12,000/mm³, less than 4000/mm³, or more than 10% immature (band) forms.

See Appendix 1 for a glossary of terms.

Description of the intervention

The main purpose of treatment is to decrease the mortality and morbidity associated with acute pancreatitis. The various pharmacological interventions that have been evaluated in the treatment of acute pancreatitis include agents such as somatostatin or octreotide that decrease pancreatic secretions; protease inhibitors such as gabexate mesilate, aprotinin, ulinastatin, and nafamostat; antioxidants such as vitamin C and selenium; platelet activating factor such as lexipafant; other agents that modulate the inflammatory pathway such as steroids and tumour necrosis factor‐alpha (TNF‐α) antibody; probiotics; and antibiotics (Bang 2008; Neumann 2011; Rada 2011; Yang 2011). We included any pharmacological intervention aimed at the treatment of acute pancreatitis.

We did not cover endoscopic sphincterotomy for the treatment of common bile duct stones (Ayub 2010), nor did we focus on endoscopic, radiology‐guided percutaneous treatments or surgical treatments for treatment of complications of acute pancreatitis (Tenner 2013). Furthermore, we did not cover the use of non‐steroidal anti‐inflammatory drugs (NSAIDs) or other drugs such as somatostatin analogues for preventing postendoscopic retrograde cholangiopancreatography (post‐ECRP)‐induced pancreatitis (Elmunzer 2012; Zhang 2009).

How the intervention might work

Somatostatin and its analogues decrease pancreatic secretion (Bang 2008). Since autodigestion (breakdown of pancreas) due to trypsinogen activation is one of the mechanisms believed to cause acute pancreatitis, decreasing pancreatic secretion can decrease the amount of trypsinogen. Inhibition of trypsin by protease inhibitors may result in decreased damage to the pancreas (Neumann 2011). Antioxidants, platelet‐activating factor inhibitors, steroids, and TNF‐α antibody are all aimed at decreasing the inflammatory response or at mitigating the damage resulting from the inflammatory response (Bang 2008). Probiotics decrease the bacterial colonisation of the gut, and antibiotics have antibacterial actions (Bang 2008).

Why it is important to do this review

Despite various pharmacological interventions being evaluated in acute pancreatitis, none is currently recommended in the treatment of acute pancreatitis, with the exception of antibiotics in infected necrotising pancreatitis (Tenner 2013). Systematic reviews and meta‐analyses increase the precision of the treatment effects (i.e. they provide a narrower range of the average treatment effect) (Higgins 2011), and so decrease the risk of a type II error (concluding that there is no difference between treatments when there is actually a difference). Systematic reviews also help in identifying the differences in the treatment effects between studies and allow exploration of the reasons behind these differences. Many studies have compared these interventions with placebo or with no treatment. It is therefore not possible to obtain accurate information on how one treatment compares with another treatment. Multiple treatment comparisons or a network meta‐analysis allow comparison of several treatments simultaneously and provide information on the relative effect of one treatment versus another, even when there is no direct comparison. There is no Cochrane Review or network meta‐analysis on this topic. So, we planned to perform a network meta‐analysis if the type of participants were included across all the comparisons. This systematic review will identify the relative effects of different treatments and identify any research gaps.

Objectives

To assess the effects of different pharmacological interventions in people with acute pancreatitis.

Methods

Criteria for considering studies for this review

Types of studies

We included only randomised controlled trials (RCTs). We included studies reported as full text, those published as abstract only, and unpublished data.

Types of participants

We included adults with acute pancreatitis irrespective of the severity (mild, moderately severe, or severe acute pancreatitis) or the type of acute pancreatitis (acute interstitial oedematous pancreatitis or necrotising pancreatitis).

Types of interventions

We included trials comparing any pharmacological interventions mentioned above with another, with placebo, or with no intervention, provided that the only difference between the randomised groups was the pharmacological intervention or interventions being assessed. Some of the interventions that we included are listed below.

Activated protein C.
Antibiotics.
Antioxidants.
Aprotinin.
Calcitonin.
Cimetidine.
EDTA (ethylenediaminetetraacetic acid).
Gabexate.
Glucagon.
Iniprol.
Lexipafant.
Octreotide.
Omeprazole.
Probiotics.
Somatostatin.
Thymosin.
Ulinastatin.

We did not combine the different somatostatin analogues (such as somatostatin or octreotide) as a single treatment in order to avoid further clinical heterogeneity. We assessed a combination of drugs as a separate treatment.

Types of outcome measures

Primary outcomes

Mortality.
1. Short‐term mortality (in‐hospital mortality or mortality within six months).
2. Long‐term mortality (at maximum follow‐up).
Serious adverse events (within six months). We accepted the definition of serious adverse events from the International Conference on Harmonisation ‐ Good Clinical Practice guideline (ICH‐GCP 1997): any untoward medical occurrence that results in death, is life‐threatening, requires inpatient hospitalisation or prolongation of existing hospitalisation, or results in persistent or significant disability/incapacity. Wealso accepted other variations of ICH‐GCP classifications such as Food and Drug Administration (FDA) classification (FDA 2006), Medicines and Healthcare products Regulatory Agency (MHRA) classification (MHRA 2013).
1. Proportion of people who developed serious adverse events (i.e. the percentage of people who developed one or more serious adverse events) and the number of serious adverse events (i.e. the total number of serious adverse events in each group regardless of the number of people in whom the serious adverse events developed).
2. Organ failure (however reported by authors).
3. Infected necrotising pancreatitis (cytology or positive culture).
4. Sepsis (however reported by authors).
Health‐related quality of life (using any validated scale).
1. Short‐term (four weeks to three months).
2. Medium‐term (three months to one year).
3. Long‐term (more than one year).
Health‐related quality of life (using any validated scale).
1. Short‐term (four weeks to three months).
2. Medium‐term (three months to one year).
3. Long‐term (more than one year).

Secondary outcomes

Adverse events (within six months). We accepted all adverse events reported by the trial authors, irrespective of the severity of the adverse event.
Measures of decreased complications and earlier recovery (within six months).
1. Length of hospital stay (including the index admission for acute pancreatitis and any disease‐related or intervention‐related readmissions including those for recurrent episodes).
2. Length of intensive care unit (ICU) stay (including the index admission for acute pancreatitis and any disease‐ or intervention‐related readmissions).
3. Requirement for additional invasive intervention such as necrosectomy for pancreatic necrosis, endoscopic or radiological drainage of collections.
4. Time to return to normal activity (return to pre‐acute pancreatitis episode mobility without any additional caregiver support).
5. Time to return to work (in those who were employed previously).
Costs (within six months).

We chose the above clinical outcomes based on the necessity to assess whether the pharmacological interventions were effective in decreasing complications, thereby decreasing the length of ICU and hospital stay, decreasing any additional interventions, and resulting in earlier return to normal activity and work as well as improvement in quality of life. The costs provide an indication of resource requirement.

We did not regard the reporting of the outcomes listed here as an inclusion criterion for the review.

Search methods for identification of studies

Electronic searches

We conducted a literature search to identify all published and unpublished randomised controlled trials. The literature search identified potential studies in all languages. We translated the non‐English language papers and fully assessed them for potential inclusion in the review as necessary.

We searched the following electronic databases for identifying potential studies.

Cochrane Central Register of Controlled Trials (CENTRAL; Issue 9, 2016; searched 7 October 2016; Appendix 2).
MEDLINE (1966 to 7 October 2016; Appendix 3).
Embase (1988 to 7 October 2016; Appendix 4).
Science Citation Index (1982 to 7 October 2016; Appendix 5).

We also conducted a search of ClinicalTrials.gov (Appendix 6) and World Health Organization International Clinical Trials Registry Platform (WHO ICTRP) (Appendix 8) on 7 October 2016.

Searching other resources

We checked the reference lists of all primary studies and review articles for additional references. We contacted authors of identified trials and asked them to identify any other published and unpublished studies.

We searched for errata or retractions from eligible trials on www.ncbi.nlm.nih.gov/pubmed on 7 October 2016.

Data collection and analysis

Selection of studies

Two review authors (KG and AB) independently screened titles and abstracts of all the potential studies that we identified through the searches and coded them as 'retrieve' (eligible or potentially eligible/unclear) or 'do not retrieve'. We retrieved the full‐text study reports, and two review authors (KG and RK or EM) independently screened them and identified studies for inclusion; we identified and recorded reasons for exclusion of the ineligible studies. We resolved any disagreement through discussion. We identified and excluded duplicates and collated multiple reports of the same study so that each study rather than each report was the unit of interest in the review. We planned to contact the investigators of trials of unclear eligibility. We recorded the selection process in sufficient detail to complete a PRISMA flow diagram (Figure 1) and a 'Characteristics of excluded studies' table.

Figure 1

Study flow diagram.

Data extraction and management

We used a standard data collection form for study characteristics and outcome data, which had been piloted on three studies in the review. Two review authors (KG and RK or EM) independently extracted the following study characteristics.

Methods: study design, total duration study and run‐in, number of study centres and location, study setting, withdrawals, date of study.
Participants: number (N), mean age, age range, sex, severity and type of acute pancreatitis, inclusion criteria, exclusion criteria.
Interventions: intervention, comparison, co‐interventions, number of participants randomised to each group.
Outcomes: primary and secondary outcomes specified and collected, time points reported. For binary outcomes, we obtained the number of participants with events and the number of participants included in the analysis in each group. For continuous outcomes, we obtained the unit or scale of measurement, mean, standard deviation, and the number of participants included in the analysis for each group. For count outcomes, we obtained the number of events and number of participants included in the analysis in each group. For time‐to‐event outcomes, we obtained the proportion of people with events, the average duration of follow‐up of participants in the trial, and the number of participants included in the analysis for each group.
Notes: funding for trial, notable conflicts of interest of trial authors.

Two review authors (KG and RK or EM) independently extracted outcome data from included studies. If outcomes were reported at multiple time points, we planned to extract the data for all time points. We obtained information on the number of participants with adverse events (or serious adverse events) and the number of such events where applicable. We planned to extract all information on costs using the currency reported by the trial authors and planned to convert this to USD at the conversion rates on the day of the analysis. We extracted data for every trial arm that was an included intervention. If studies reported outcome data in an unusable way, we attempted to contact the trial authors and tried to obtain usable data. If we were unable to obtain usable data despite this, we planned to summarise the unusable data in an appendix. We resolved disagreements by consensus. One review author (EM) copied across the data for 'Characteristics of included studies' and 'Characteristics of excluded studies' from the data collection form into the Review Manager 5 (RevMan 5) file (RevMan 2014). One review author (KG) copied across the data for 'Data and analyses' from the data collection form into the RevMan 5 file. We double‐checked that the data were entered correctly by comparing the study reports with how the data were presented in the systematic review.

Assessment of risk of bias in included studies

Two review authors (KG and RK or EM) independently assessed the risk of bias for each study using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We resolved any disagreements by discussion. We assessed the risk of bias according to the following domains.

Random sequence generation.
Allocation concealment.
Blinding of participants and personnel.
Blinding of outcome assessment.
Incomplete outcome data.
Selective outcome reporting.
Bias due to funding source.
Other potential bias.

We graded each potential source of bias as high, low, or unclear and provided a quote from the study report together with a justification for our judgement in the 'Risk of bias' tables. We summarised the risk of bias judgements across different studies for each of the domains listed. We considered blinding separately for different key outcomes where necessary, for example, for unblinded outcome assessment, risk of bias for all‐cause mortality may be very different than for a participant‐reported pain scale. Where information on risk of bias relates to unpublished data or to correspondence with a trial author, we planned to note this in the 'Risk of bias' table. We presented the risk of bias in each pair‐wise comparison in Table 1.

Open in table viewer

Table 1. Characteristics of included studies (ordered by comparisons)

Study name	No of participants randomised	Postrandomisation dropouts	No of participants for whom outcome was reported	Treatment 1	Treatment 2	Selection bias	Performance and detection bias	Attrition bias	Selective reporting bias	Other bias
Pettila 2010	32	0	32	Activated protein C	Placebo	Unclear	Low	Low	High	High
Barreda 2009	80	22	58	Antibiotics	No active intervention	Unclear	Unclear	High	Low	Unclear
Delcenserie 1996	23	0	23	Antibiotics	No active intervention	Unclear	Unclear	Low	Low	Unclear
Delcenserie 2001	81	Not stated	81	Antibiotics	No active intervention	Unclear	Unclear	Unclear	Low	Unclear
Dellinger 2007	100	0	100	Antibiotics	Placebo	Low	Low	Low	Low	High
Finch 1976	62	4	58	Antibiotics	No active intervention	Unclear	Unclear	High	Low	Unclear
Garcia‐Barrasa 2009	46	5	41	Antibiotics	Placebo	Unclear	Low	High	Low	Low
Hejtmankova 2003	41	Not stated	41	Antibiotics	No active intervention	Unclear	Unclear	Unclear	Low	Unclear
Isenmann 2004	119	5	114	Antibiotics	Placebo	Unclear	Low	High	High	High
Llukacaj 2012	80	Not stated	80	Antibiotics	Placebo	Unclear	Low	Unclear	High	Unclear
Luiten 1995	109	7	102	Antibiotics	No active intervention	Unclear	Unclear	High	Low	Unclear
Nordback 2001	90	32	58	Antibiotics	Placebo	Unclear	Unclear	High	Low	Unclear
Poropat 2015	47	0	47	Antibiotics	No active intervention	Unclear	Unclear	Low	Low	Unclear
Pederzoli 1993a	74	Not stated	74	Antibiotics	No active intervention	Unclear	Unclear	Low	Low	Unclear
Rokke 2007	73	0	73	Antibiotics	No active intervention	Unclear	High	Low	Low	High
Sainio 1995	60	0	60	Antibiotics	No active intervention	Unclear	Unclear	Low	Low	Unclear
Spicak 2002	63	Not stated	63	Antibiotics	No active intervention	Unclear	Unclear	Unclear	Low	Unclear
Spicak 2003	41	Not stated	41	Antibiotics	No active intervention	Unclear	Unclear	Unclear	Low	Unclear
Xue 2009	59	3	56	Antibiotics	No active intervention	Unclear	Unclear	High	Low	Low
Bansal 2011	44	5	39	Antioxidants	No active intervention	Unclear	High	High	Low	Low
Birk 1994	20	Not stated	20	Antioxidants	No active intervention	Unclear	Unclear	Unclear	High	Unclear
Marek 1999	73	0	73	Antioxidants	Placebo	Unclear	Unclear	Low	High	Unclear
Sateesh 2009	56	3	53	Antioxidants	No active intervention	Unclear	High	High	Low	Unclear
Siriwardena 2007	43	0	43	Antioxidants	Placebo	Low	Low	Low	Low	High
Vege 2015	28	Not stated	28	Antioxidants	Placebo	Unclear	Low	Low	Low	Unclear
Chooklin 2007	34	Not stated	34	Antioxidants plus Corticosteroids	No active intervention	Unclear	Unclear	Unclear	High	Unclear
MRC Multicentre Trial 1977 (this is a 3‐armed trial; the numbers stated included all 3 arms)	264	7	257	Aprotinin	Placebo	Unclear	Low	High	High	High
Balldin 1983	55	Not stated	55	Aprotinin	No active intervention	Unclear	Unclear	Unclear	Low	High
Berling 1994	48	Not stated	48	Aprotinin	No active intervention	Unclear	Low	Low	Low	High
Imrie 1978	161	Not stated	161	Aprotinin	Placebo	Unclear	Low	Unclear	Low	High
Imrie 1980	50	Not stated	50	Aprotinin	Placebo	Unclear	Low	Unclear	High	Unclear
Storck 1968	43	Not stated	43	Aprotinin	Placebo	Unclear	Low	Unclear	High	Unclear
Trapnell 1974	105	Not stated	105	Aprotinin	Placebo	Low	Low	Unclear	High	High
MRC Multicentre Trial 1977 (this is a 3‐armed trial; the numbers stated included all 3 arms)	264	7	257	Aprotinin	Glucagon	Unclear	Low	High	High	High
Goebell 1979	94	Not stated	94	Calcitonin	Placebo	Unclear	Low	Unclear	Low	Unclear
Martinez 1984	31	0	31	Calcitonin	Placebo	Unclear	Unclear	Low	High	Unclear
Perezdeoteyza 1980	40	Not stated	40	Cimetidine	Placebo	Unclear	Low	Unclear	High	Unclear
Sillero 1981	60	Not stated	60	Cimetidine	Placebo	Low	Unclear	Unclear	High	Unclear
Tykka 1985	64	0	64	EDTA	Placebo	Unclear	Low	Low	Low	High
Frulloni 1994	116	Not stated	116	Gabexate	Aprotinin	Unclear	Unclear	Unclear	Low	Unclear
Pederzoli 1993b	199	17	182	Gabexate	Aprotinin	Unclear	Low	High	Low	Unclear
Buchler 1993	223	Not stated	223	Gabexate	Placebo	Low	Low	Low	Low	Unclear
Chen 2000	52	Not stated	52	Gabexate	Placebo	Unclear	Unclear	Unclear	Low	Unclear
Freise 1986	50	Not stated	50	Gabexate	Placebo	Unclear	Low	Unclear	Low	Unclear
Goebell 1988	162	11	151	Gabexate	Placebo	Unclear	Low	High	Low	Unclear
Valderrama 1992	105	5	100	Gabexate	Placebo	Low	Low	High	Low	High
Kirsch 1978	150	Not stated	150	Glucagon	Atropine	Unclear	Unclear	Unclear	Low	Unclear
MRC Multicentre Trial 1977 (this is a 3‐armed trial; the numbers stated included all 3 arms)	264	7	257	Glucagon	Placebo	Unclear	Unclear	Unclear	Low	High
Debas 1980	66	Not stated	66	Glucagon	Placebo	Unclear	Low	Unclear	Low	Unclear
Dürr 1978	69	Not stated	69	Glucagon	Placebo	Unclear	Low	Unclear	High	Unclear
Kalima 1980	80	9	71	Glucagon	Placebo	Unclear	Unclear	High	Low	Unclear
Kronborg 1980	22	Not stated	22	Glucagon	Placebo	Unclear	Low	Unclear	High	Unclear
Gilsanz 1978	62	Not stated	62	Glucagon	Oxyphenonium	Unclear	Low	Unclear	Low	Unclear
Hansky 1969	24	Not stated	24	Iniprol	No active intervention	Unclear	High	Unclear	High	High
Johnson 2001	291	1	290	Lexipafant	Placebo	Unclear	Low	High	Low	High
Kingsnorth 1995	83	Not stated	83	Lexipafant	Placebo	Unclear	Low	Unclear	High	High
McKay 1997b	51	1	50	Lexipafant	Placebo	Unclear	Low	High	High	High
Bredkjaer 1988	66	9	57	NSAID	Placebo	Unclear	Unclear	Unclear	High	Unclear
Ebbehøj 1985	30	0	30	NSAID	Placebo	Unclear	Low	Low	High	High
McKay 1997a	58	0	58	Octreotide	Placebo	Low	Low	Low	Low	Unclear
Ohair 1993	180	Not stated	180	Octreotide	Placebo	Unclear	Unclear	Unclear	High	Unclear
Paran 1995	51	13	38	Octreotide	No active intervention	Unclear	High	High	Low	Unclear
Uhl 1999	302	0	302	Octreotide	Placebo	Unclear	Low	Low	Low	High
Wang 2013c	372	Not stated	372	Octreotide	No active intervention	Unclear	Unclear	High	Low	Low
Yang 2012	163	6	157	Octreotide	No active intervention	Unclear	Unclear	High	High	Low
Wang 2013b	354	Not stated	354	Octreotide plus NSAID	Octreotide	Unclear	Unclear	Unclear	High	Unclear
Guo 2015	120	Not stated	120	Octreotide plus ulinastatin	Octreotide	Unclear	Unclear	Unclear	Low	Unclear
Besselink 2008	298	2	296	Probiotics	Placebo	Low	Low	High	Low	High
Olah 2007	83	21	62	Probiotics	No active intervention	Unclear	Low	High	High	Unclear
Plaudis 2010	90	Not stated	58	Probiotics	No active intervention	Unclear	Low	Unclear	High	Unclear
Sharma 2011	50	0	50	Probiotics	Placebo	Unclear	Low	Low	High	High
Zhu 2014	39	Not stated	39	Probiotics	Placebo	Unclear	Low	Unclear	High	Unclear
Grupo Español 1996	70	9	61	Somatostatin	Placebo	Unclear	Low	High	High	Unclear
Choi 1989	71	Not stated	71	Somatostatin	No active intervention	Unclear	Unclear	Unclear	Low	Unclear
Gjørup 1992	63	Not stated	63	Somatostatin	Placebo	Unclear	Low	Unclear	Low	Unclear
Luengo 1994	100	Not stated	100	Somatostatin	No active intervention	Unclear	Low	Unclear	High	Unclear
Moreau 1986	87	3	84	Somatostatin	Placebo	Unclear	Low	Unclear	High	High
Usadel 1985	77	Not stated	77	Somatostatin	Placebo	Unclear	Low	Unclear	High	Unclear
Wang 2013a (this is a 3‐armed trial; the numbers stated included all 3 arms)	183	Not stated	183	Somatostatin	No active intervention	Unclear	Low	Unclear	Low	Low
Yang 1999	48	Not stated	48	Somatostatin	No active intervention	Unclear	Unclear	Unclear	High	Unclear
Xia 2014	140	Not stated	140	Somatostatin plus omeprazole	No active intervention	Unclear	Unclear	Unclear	Low	Unclear
Wang 2013a (this is a 3‐armed trial; the numbers stated included all 3 arms)	183	Not stated	183	Somatostatin plus ulinastatin	Placebo	Unclear	Unclear	Unclear	High	Unclear
Wang 2013a (this is a 3‐armed trial; the numbers stated included all 3 arms)	183	Not stated	183	Somatostatin plus ulinastatin	Somatostatin	Unclear	Low	Unclear	Low	Low
Wang 2016 (this is a 4‐armed trial; the numbers stated included all 4 arms)	492	0	492	Somatostatin plus ulinastatin	Somatostatin	Low	Low	Low	Low	Low
Wang 2016 (this is a 4‐armed trial; the numbers stated included all 4 arms)	492	0	492	Somatostatin plus gabexate	Somatostatin	Low	Low	Low	Low	Low
Wang 2016 (this is a 4‐armed trial; the numbers stated included all 4 arms)	492	0	492	Somatostatin plus ulinastatin plus gabexate	Somatostatin	Low	Low	Low	Low	Low
Wang 2016 (this is a 4‐armed trial; the numbers stated included all 4 arms)	492	0	492	Somatostatin plus ulinastatin	Somatostatin plus gabexate	Low	Low	Low	Low	Low
Wang 2016 (this is a 4‐armed trial; the numbers stated included all 4 arms)	492	0	492	Somatostatin plus ulinastatin plus gabexate	Somatostatin plus gabexate	Low	Low	Low	Low	Low
Wang 2016 (this is a 4‐armed trial; the numbers stated included all 4 arms)	492	0	492	Somatostatin plus ulinastatin plus gabexate	Somatostatin plus ulinastatin	Low	Low	Low	Low	Low
Wang 2011	24	Not stated	24	Thymosin	Placebo	Unclear	Low	Unclear	High	Unclear
Abraham 2013	135	6	129	Ulinastatin	Placebo	Unclear	Low	High	Low	Unclear
Chen 2002a	68	6	62	Ulinastatin	Gabexate	Unclear	Unclear	High	High	Unclear
Chen 2002b	26	1	25	Ulinastatin	Octreotide	Unclear	Unclear	High	High	Unclear

When considering treatment effects, we took into account the risk of bias for the studies that contribute to that outcome by a sensitivity analysis.

Assessment of bias in conducting the systematic review

We conducted the review according to the published protocol and reported any deviations from it in the 'Differences between protocol and review' section of this review.

Measures of treatment effect

For dichotomous variables (short‐term mortality, proportion of participants with adverse events, requirement for additional interventions), we calculated the odds ratio (OR) with 95% confidence interval (CI). For continuous variables, such as length of hospital stay, ICU stay, time to return to normal activity, time to return to work, and costs, we planned to calculate the mean difference (MD) with 95% CI. We planned to use standardised mean difference (SMD) with 95% CI for quality of life if different scales were used. For count outcomes such as the number of adverse events, we calculated the rate ratio with 95% CIs. For time‐to‐event data, such as long‐term mortality, we planned to use the hazard ratio (HR) with a 95% CI. However, only one trial reported mortality beyond 3 months and presented the number of deaths at two years. We analysed this information as binary data.

A common way that trial authors indicate when they have skewed data is by reporting medians and interquartile ranges. When we encountered this, we reported the difference in means or medians in a table.

Unit of analysis issues

The unit of analysis was individual participants with acute pancreatitis. As anticipated, we did not find any cluster‐randomised trials for this comparison.

In multi‐arm trials, the models account for the correlation between trial‐specific treatment effects from the same trial.

Dealing with missing data

We attempted to contact investigators or study sponsors in order to verify key study characteristics and obtain missing numerical outcome data where possible (e.g. when a study was identified as abstract only). For binary, count, and time‐to‐event outcomes, we performed an intention‐to‐treat analysis whenever possible (Newell 1992). Since this was not possible, we performed an available‐case analysis but planned to assess the impact of 'best‐best', 'best‐worst', 'worst‐best', and 'worst‐worst' scenario analyses on the results for binary outcomes. For continuous outcomes, we planned to perform an available‐case analysis. If we were unable to obtain the information from the investigators or study sponsors, we planned to impute the mean from the median (i.e. consider the median as the mean) and the standard deviation from the standard error, interquartile range, or P values according to the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011), but we planned to assess the impact of including such studies as indicated in a sensitivity analysis. If we were unable to calculate the standard deviation from the standard error, interquartile range, or P values, we planned to impute the standard deviation as the highest standard deviation in the remaining trials included in the outcome, being fully aware that this method of imputation would decrease the weight of the studies in the meta‐analysis of mean difference and shift the effect estimate towards no effect for standardised mean difference. We planned to assess the impact of including such studies by sensitivity analysis.

Assessment of heterogeneity

We assessed the heterogeneity in each pair‐wise comparison by assessing the Higgins I² (Higgins 2003), the Chi² test with significance set at a P value less than 0.10, and by visual inspection.

Assessment of reporting biases

We attempted to contact trial authors, asking them to provide missing outcome data. Where this was not possible, and if we thought that the missing data may introduce serious bias, we planned to explore the impact of including such studies in the overall assessment of results by a sensitivity analysis.

If we were able to pool more than 10 trials for a specific comparison, we created and examined a funnel plot to explore possible publication biases. We used Egger's test to determine the statistical significance of the reporting bias (Egger 1997). We considered a P value of less than 0.05 to indicate statistically significant reporting bias.

Data synthesis

We undertook meta‐analyses only where this was meaningful (i.e. if the treatments, participants and the underlying clinical question were similar enough for pooling to make sense). In general, we favoured performing a meta‐analysis and clearly highlighted the reason for not performing one if we decided against it. We used both the fixed‐effect and random‐effects model, reporting the fixed‐effect model when the choice of models did not alter the conclusion and the random‐effects model when it did. We did not perform a network meta‐analysis as planned because of the lack of information on potential effect modifiers and differences of type of participants included in the different comparisons, when information was available (i.e. the transitivity assumption was not satisfied).

Subgroup analysis and investigation of heterogeneity

We planned to perform the following subgroup analyses regardless of heterogeneity.

Different types of acute pancreatitis (acute interstitial oedematous pancreatitis or necrotising pancreatitis).
Different severity of acute pancreatitis (mild pancreatitis versus moderate or severe acute pancreatitis).
Presence of persistent organ failure (mild or moderate acute pancreatitis versus severe acute pancreatitis).
Presence of infection (infected necrotising pancreatitis versus non‐infected necrotising pancreatitis).

We planned to calculate the test for subgroup differences to identify differences between subgroups.

Sensitivity analysis

We planned to perform the following sensitivity analyses defined a priori to assess the robustness of our conclusions.

Excluding trials at unclear or high risk of bias (one or more of the 'Risk of bias' domains classified as unclear or high).
Excluding trials in which either the mean or the standard deviation or both were imputed.
Imputation of binary outcomes under 'best‐best', 'best‐worst', 'worst‐best', and 'worst‐worst' scenarios.

'Summary of findings' table

Although we planned to create a 'Summary of findings' table using all the outcomes, this would have resulted in a incomprehensible table. So, we presented the 'Summary of findings' table for the primary outcomes only. We used the five GRADE considerations (study limitations, inconsistency of effect, imprecision, indirectness and publication bias) to assess the quality of the body of evidence as it related to the studies contributing data to the meta‐analyses for the prespecified outcomes. We justified all decisions to down‐ or upgrade the quality rating of studies using footnotes, making comments to aid the reader's understanding of the review where necessary. We considered whether there was any additional outcome information that we were not able to incorporate into meta‐analyses and planned to note this in the comments, stating whether it supported or contradicted the information from the meta‐analyses.

Reaching conclusions

We based our conclusions only on findings from the quantitative or narrative synthesis of included studies for this review. We have avoided making recommendations for practice, and our implications for research give the reader a clear sense of where the focus of any future research in the area should be and what the remaining uncertainties are.

Results

Description of studies

Results of the search

We identified a total of 13,825 references through electronic searches of CENTRAL (1345 records), MEDLINE (5649 records), Embase (4102 records), Science Citation Index Expanded (2604 records), World Health Organization International Clinical Trials Registry Platform (78 records) and ClinicalTrials.gov (47 records). After removing 3548 duplicates, we obtained 10,277 references. We then excluded 10,030 clearly irrelevant references through screening titles and reading abstracts. We sought 247 references for further assessment but could not obtain 2 (Hansen 1966; Perez 1980). Seven references were ongoing trials, suspended trials, or completed trials identified from clinical registers with no interim reports available (ChiCTR‐IPR‐16008301; EUCTR2014‐004844‐37‐ES; NCT01132521; NCT02025049; NCT02212392; NCT02692391; NCT02885441). We did not identify any new trials by scanning reference lists of the identified randomised trials. We excluded 102 references for the reasons listed under the table 'Characteristics of excluded studies'. In total, 136 references (84 trials) met the inclusion criteria. The reference flow is summarised in the study flow diagram (Figure 1).

Included studies

A total of 8234 participants were included in these 84 trials. Six trials (N = 658) did not report any of the outcomes of interest for this review (Birk 1994; Chooklin 2007; Marek 1999; Moreau 1986; Plaudis 2010; Wang 2013b). The remaining 78 trials excluded 210 participants after randomisation. Thus, a total of 7366 participants in 78 trials contributed to one or more outcomes for this review.

One trial included only participants with acute interstitial oedematous pancreatitis (Chen 2002a); 12 trials included only participants with acute necrotising pancreatitis (Barreda 2009; Chen 2002b; Delcenserie 2001; Dellinger 2007; Frulloni 1994; Garcia‐Barrasa 2009; Llukacaj 2012; Nordback 2001; Pederzoli 1993a; Rokke 2007; Sainio 1995; Xue 2009); the remaining trials did not state clearly whether they included any participants with acute necrotising pancreatitis. All the trials that included acute necrotising pancreatitis either stated explicitly or implied that they excluded participants with infected necrotising pancreatitis.

Two trials included only participants with mild acute pancreatitis (Chen 2002a; Yang 2012). Twenty‐six trials included only severe acute pancreatitis (Balldin 1983; Berling 1994; Birk 1994; Chen 2000; Chen 2002b; Chooklin 2007; Delcenserie 1996; Dellinger 2007; Garcia‐Barrasa 2009; Grupo Español 1996; Guo 2015; Hejtmankova 2003; Luiten 1995; Martinez 1984; Olah 2007; Pettila 2010; Plaudis 2010; Rokke 2007; Spicak 2002; Spicak 2003; Wang 2011; Wang 2013a; Wang 2016; Xia 2014; Xue 2009; Zhu 2014). Two trials reported data separately for mild and severe acute pancreatitis (Abraham 2013; Wang 2013c). These trials presented the data separately for mild pancreatitis and acute severe pancreatitis. The remaining trials either included mild and severe acute pancreatitis or did not state the severity of pancreatitis in the participants. It should be noted that none of the trials used the current definition of severe acute pancreatitis (i.e. organ failure persisting for 48 hours or more).

The potential effect modifiers, arranged by comparisons, are shown in Table 2. As shown in the table, important potential effect modifiers were missing. In addition, it appeared that most trials in the comparison on antibiotics versus no active intervention included participants with necrotising pancreatitis. Because of this, there were serious concerns about the inclusion of similar participants in the different comparisons.

Open in table viewer

Table 2. Potential effect modifiers (ordered by comparisons)

Study name	Treatment 1	Treatment 2	Severe pancreatitis	Necrotising pancreatitis	Organ failure	Infection
Pettila 2010	Activated protein C	Placebo	yes	not stated	not stated	not stated
Barreda 2009	Antibiotics	No active intervention	not stated	yes	not stated	not stated
Delcenserie 1996	Antibiotics	No active intervention	yes	not stated	not stated	not stated
Delcenserie 2001	Antibiotics	No active intervention	not stated	yes	not stated	not stated
Dellinger 2007	Antibiotics	Placebo	yes	yes	not stated	no
Finch 1976	Antibiotics	No active intervention	not stated	not stated	not stated	not stated
Garcia‐Barrasa 2009	Antibiotics	Placebo	yes	yes	not stated	not stated
Hejtmankova 2003	Antibiotics	No active intervention	yes	not stated	not stated	not stated
Isenmann 2004	Antibiotics	Placebo	not stated	not stated	not stated	not stated
Llukacaj 2012	Antibiotics	Placebo	not stated	yes	not stated	no
Luiten 1995	Antibiotics	No active intervention	yes	not stated	not stated	no
Nordback 2001	Antibiotics	Placebo	not stated	yes	no	not stated
Pederzoli 1993a	Antibiotics	No active intervention	not stated	yes	not stated	not stated
Rokke 2007	Antibiotics	No active intervention	yes	yes	not stated	not stated
Sainio 1995	Antibiotics	No active intervention	not stated	yes	not stated	not stated
Spicak 2002	Antibiotics	No active intervention	yes	not stated	not stated	not stated
Spicak 2003	Antibiotics	No active intervention	yes	not stated	not stated	not stated
Xue 2009	Antibiotics	No active intervention	yes	yes	not stated	no
Bansal 2011	Antioxidants	No active intervention	not stated	not stated	not stated	not stated
Birk 1994	Antioxidants	No active intervention	yes	not stated	not stated	not stated
Marek 1999	Antioxidants	Placebo	not stated	not stated	not stated	not stated
Sateesh 2009	Antioxidants	No active intervention	not stated	not stated	not stated	not stated
Siriwardena 2007	Antioxidants	Placebo	not stated	not stated	not stated	not stated
Vege 2015	Antioxidants	Placebo	not stated	not stated	not stated	not stated
Chooklin 2007	Antioxidants plus corticosteroids	No active intervention	yes	not stated	not stated	not stated
Balldin 1983	Aprotinin	No active intervention	yes	not stated	not stated	not stated
Berling 1994	Aprotinin	No active intervention	yes	not stated	not stated	not stated
Imrie 1978	Aprotinin	Placebo	not stated	not stated	not stated	not stated
Imrie 1980	Aprotinin	Placebo	not stated	not stated	not stated	not stated
MRC Multicentre Trial 1977	Aprotinin	Placebo	not stated	not stated	not stated	not stated
Storck 1968	Aprotinin	Placebo	not stated	not stated	not stated	not stated
Trapnell 1974	Aprotinin	Placebo	not stated	not stated	not stated	not stated
Goebell 1979	Calcitonin	Placebo	not stated	not stated	not stated	not stated
Martinez 1984	Calcitonin	Placebo	yes	not stated	not stated	not stated
Perezdeoteyza 1980	Cimetidine	Placebo	not stated	not stated	not stated	not stated
Sillero 1981	Cimetidine	Placebo	not stated	not stated	not stated	not stated
Tykka 1985	EDTA	Placebo	not stated	not stated	not stated	not stated
Buchler 1993	Gabexate	Placebo	not stated	not stated	not stated	not stated
Chen 2000	Gabexate	Placebo	yes	not stated	yes	not stated
Freise 1986	Gabexate	Placebo	not stated	not stated	not stated	not stated
Goebell 1988	Gabexate	Placebo	not stated	not stated	not stated	not stated
Valderrama 1992	Gabexate	Placebo	not stated	not stated	not stated	not stated
Debas 1980	Glucagon	Placebo	not stated	not stated	not stated	not stated
Dürr 1978	Glucagon	Placebo	not stated	not stated	not stated	not stated
Kalima 1980	Glucagon	Placebo	not stated	not stated	not stated	not stated
Kronborg 1980	Glucagon	Placebo	not stated	not stated	not stated	not stated
MRC Multicentre Trial 1977	Glucagon	Placebo	not stated	not stated	not stated	not stated
Hansky 1969	Iniprol	No active intervention	not stated	not stated	not stated	not stated
Johnson 2001	Lexipafant	Placebo	not stated	not stated	not stated	not stated
Kingsnorth 1995	Lexipafant	Placebo	not stated	not stated	not stated	not stated
McKay 1997b	Lexipafant	Placebo	not stated	not stated	not stated	not stated
Bredkjaer 1988	NSAID	Placebo	not stated	not stated	not stated	not stated
Ebbehøj 1985	NSAID	Placebo	not stated	not stated	not stated	not stated
McKay 1997b	Octreotide	Placebo	not stated	not stated	not stated	not stated
Ohair 1993	Octreotide	Placebo	not stated	not stated	not stated	not stated
Paran 1995	Octreotide	No active intervention	not stated	not stated	not stated	not stated
Uhl 1999	Octreotide	Placebo	not stated	not stated	not stated	not stated
Wang 2013c (mild pancreatitis)	Octreotide	No active intervention	no	not stated	not stated	not stated
Wang 2013c (severe pancreatitis)	Octreotide	No active intervention	yes	not stated	not stated	not stated
Yang 2012	Octreotide	No active intervention	no	not stated	not stated	not stated
Besselink 2008	Probiotics	Placebo	not stated	not stated	not stated	not stated
Olah 2007	Probiotics	No active intervention	yes	not stated	not stated	not stated
Plaudis 2010	Probiotics	No active intervention	yes	not stated	not stated	not stated
Sharma 2011	Probiotics	Placebo	not stated	not stated	not stated	not stated
Zhu 2014	Probiotics	Placebo	yes	not stated	not stated	not stated
Choi 1989	Somatostatin	No active intervention	not stated	not stated	not stated	not stated
Gjørup 1992	Somatostatin	Placebo	not stated	not stated	not stated	not stated
Grupo Español 1996	Somatostatin	Placebo	yes	not stated	not stated	not stated
Luengo 1994	Somatostatin	No active intervention	not stated	not stated	not stated	not stated
Moreau 1986	Somatostatin	Placebo	not stated	not stated	not stated	not stated
Usadel 1985	Somatostatin	Placebo	not stated	not stated	not stated	not stated
Wang 2013a	Somatostatin	No active intervention	yes	not stated	not stated	not stated
Yang 1999	Somatostatin	No active intervention	not stated	not stated	not stated	not stated
Xia 2014	Somatostatin plus omeprazole	No active intervention	yes	not stated	not stated	not stated
Wang 2013a	Somatostatin plus ulinastatin	No active intervention	yes	not stated	not stated	not stated
Wang 2011	Thymosin	Placebo	yes	not stated	not stated	not stated
Abraham 2013 (mild pancreatitis)	Ulinastatin	Placebo	no	not stated	not stated	no
Abraham 2013 (severe pancreatitis)	Ulinastatin	Placebo	yes	not stated	not stated	not stated
Frulloni 1994	Gabexate	Aprotinin	not stated	yes	not stated	not stated
Pederzoli 1993b	Gabexate	Aprotinin	not stated	not stated	not stated	not stated
Kirsch 1978	Glucagon	Atropine	not stated	not stated	not stated	not stated
Chen 2002a	Ulinastatin	Gabexate	no	no	no	not stated
MRC Multicentre Trial 1977	Aprotinin	Glucagon	not stated	not stated	not stated	not stated
Guo 2015	Octerotide plus ulinastatin	Octreotide	yes	not stated	not stated	not stated
Wang 2013b	Octreotide plus NSAID	Octreotide	not stated	not stated	not stated	not stated
Chen 2002b	Ulinastatin	Octreotide	yes	yes	not stated	not stated
Gilsanz 1978	Glucagon	Oxyphenonium	not stated	not stated	not stated	not stated
Poropat 2015	Antibiotics	No active intervention	not stated	not stated	not stated	no
Wang 2016	Somatostatin plus gabexate	Somatostatin	yes	not stated	not stated	not stated
Wang 2013a	Somatostatin plus ulinastatin	Somatostatin	yes	not stated	not stated	not stated
Wang 2016	Somatostatin plus ulinastatin	Somatostatin	yes	not stated	not stated	not stated
Wang 2016	Somatostatin plus ulinastatin plus gabexate	Somatostatin	yes	not stated	not stated	not stated
Wang 2016	Somatostatin plus ulinastatin	Somatostatin plus gabexate	yes	not stated	not stated	not stated
Wang 2016	Somatostatin plus ulinastatin plus gabexate	Somatostatin plus gabexate	yes	not stated	not stated	not stated
Wang 2016	Somatostatin plus ulinastatin plus gabexate	Somatostatin plus ulinastatin	yes	not stated	not stated	not stated

Source of funding: seven trials were not funded or they were funded by agencies without vested interest in results (Bansal 2011; Garcia‐Barrasa 2009; Wang 2013a; Wang 2013c; Wang 2016; Xue 2009; Yang 2012). Pharmaceutical companies partially or fully funded 21 trials (Balldin 1983; Berling 1994; Besselink 2008; Dellinger 2007; Ebbehøj 1985; Hansky 1969; Imrie 1978; Isenmann 2004; Johnson 2001; Kingsnorth 1995; McKay 1997b; Moreau 1986; MRC Multicentre Trial 1977; Pettila 2010; Rokke 2007; Sharma 2011; Siriwardena 2007; Trapnell 1974; Tykka 1985; Uhl 1999; Valderrama 1992). The source of funding was not available from the remaining trials.

Excluded studies

None of the excluded studies were eligible for this review. The reasons for exclusion are listed in 'Characteristics of excluded studies'.

Risk of bias in included studies

We summarised the overall risk of bias in Figure 2 and Figure 3. Only Wang 2016 was at low risk of bias in all the domains and can be considered a trial at overall low risk of bias.

Figure 2

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Figure 3

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

No other bias was noted in any of the trials.