Network meta‐analysis comparing techniques and outcomes of stump closure after distal pancreatectomy

• Published in: British journal of surgery Vol. 106; no. 12; pp. 1580 - 1589
• Main Authors: Ratnayake, C. B. B., Wells, C., Hammond, J., French, J. J., Windsor, J. A., Pandanaboyana, S.
• Format: Journal Article
• Language: English
• Published: Chichester, UK John Wiley & Sons, Ltd 01-11-2019; Oxford University Press
• Subjects: Abdominal Abscess - etiology; Blood Loss, Surgical; Fistula; Hospital Mortality; Humans; Meta-analysis; Network Meta-Analysis; Operative Time; Pancreas; Pancreatectomy; Pancreatectomy - adverse effects; Pancreatectomy - methods; Pancreatic Fistula - prevention & control; Postoperative Complications - prevention & control; Surgical outcomes; Surgical Stapling; Surgical techniques; Suture Techniques; Wound Closure Techniques - adverse effects
• ISSN: 0007-1323; 1365-2168
• DOI: 10.1002/bjs.11291

Background
The incidence of postoperative pancreatic fistula (POPF) after distal pancreatectomy remains high, and different pancreatic stump closure techniques have been used to reduce the incidence. A network meta‐analysis was undertaken to compare the most frequently performed pancreatic stump closure techniques after distal pancreatectomy and determine the technique associated with the lowest POPF rate.
Methods
A systematic search of the Scopus, PubMed, MEDLINE and Embase databases was conducted to identify eligible RCTs. The primary outcome was the occurrence of clinically relevant POPF. Secondary outcomes were duration of operation, blood loss, intrabdominal collections, postoperative complications and 30‐day mortality.
Results
Sixteen RCTs including 1984 patients and eight different pancreatic stump closure techniques were included in the network meta‐analysis. Patch coverage of the pancreatic stump (round ligament or seromuscular patch) after stapler or suture closure ranked best, with the lowest rates of clinically relevant POPF, lowest volume of intraoperative blood loss, fewer intra‐abdominal abscesses, and lower rates of overall complications and 30‐day mortality. Round ligament patch closure outperformed seromuscular patch closure in preventing clinically relevant POPF with a significantly larger cohort for comparative analysis. Pancreaticoenteric anastomotic closure consistently ranked poorly for most reported postoperative outcomes.
Conclusion
Patch coverage after stapler or suture closure has the lowest POPF rate and best outcomes among stump closure techniques after distal pancreatectomy.
Antecedentes
La incidencia de fístula pancreática (postoperative pancreatic fistula, POPF) tras una pancreatectomía distal sigue siendo alta y se han utilizado diferentes técnicas para el cierre del muñón pancreático con la intención de reducir su incidencia. Se realizó un metaanálisis en red para comparar las técnicas de cierre del muñón pancreático realizadas con más frecuencia después de la pancreatectomía distal y determinar qué técnica se asocia a una menor tasa de POPF.
Métodos
Se realizó una búsqueda sistemática en las bases de datos Scopus, PubMed, Medline y EMBASE de los RCTs que podían ser incluidos en estudio. La variable principal fue la aparición de POPF clínicamente relevante. Las variables secundarias fueron el tiempo operatorio, la pérdida de sangre, las colecciones intraabdominales, las complicaciones postoperatorias y la mortalidad a los 30 días.
Resultados
En el metaanálisis se incluyeron 16 RCTs con 1.984 pacientes y 8 técnicas diferentes de cierre del muñón pancreático. Los mejores resultados (menor tasa de POPF clínicamente relevante, menor pérdida sanguínea intraoperatoria, menor número de abscesos intraabdominales, menor tasa de complicaciones generales y menor mortalidad a los 30 días) se obtuvieron con el refuerzo del muñón pancreático con parches (de ligamento redondo o seromuscular), seguidos del grapado quirúrgico o la sutura simple. El cierre con parche del ligamento redondo fue superior al parche seromuscular en la prevención de POPF clínicamente relevante en una cohorte suficiente para el análisis estadístico comparativo. Los peores resultados en la mayoría de las variables postoperatorias se obtuvieron con el cierre simple.
Conclusión
En la pancreatectomía distal, la menor tasa de POPF y los mejores resultados perioperatorios se obtuvieron con el refuerzo con parches, seguidos del grapado quirúrgico o la sutura simple.
This network meta‐analysis comparing eight different methods of stump closure following distal pancreatectomy found that patch coverage following suture or stapled closure consistently ranked the best with regard to postoperative outcomes. Conversely, anastomotic closure consistently ranked poorly in most postoperative outcomes. In contrast to multiple previous reviews that compared the pancreatic stump closure techniques available, the present study used network meta‐analysis methodology to perform multiple comparisons simultaneously while maintaining randomization.
Identifies optimal method