When to Operate After Failed Nonoperative Management




© Springer International Publishing Switzerland 2016
Abe Fingerhut, Ari Leppäniemi, Raul Coimbra, Andrew B. Peitzman, Thomas M. Scalea and Eric J. Voiglio (eds.)Emergency Surgery Course (ESC®) Manual10.1007/978-3-319-21338-5_6


6. When to Operate After Failed Nonoperative Management



Gregory A. Watson  and Andrew B. Peitzman 


(1)
Department of Surgery, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA

(2)
Mark M. Ravitch Professor and Vice-Chair Chief, Division of General Surgery, University of Pittsburgh, Pittsburgh, PA, USA

 



 

Gregory A. Watson



 

Andrew B. Peitzman (Corresponding author)





6.1 Introduction


Initial nonoperative management of patients with acute pathology is commonplace for several disorders. Inherent in this decision is the belief that surgery is best performed in a delayed fashion (when conditions are more favorable, both for the patient and the surgeon) or that surgery can be avoided altogether. However, despite our best intentions, nonoperative management will fail in a certain subset of patients initially believed to benefit from such an approach. In this chapter, we will discuss when to consider operative management (and, consequently, how to recognize that nonoperative management has failed) for several common conditions seen by general and acute care surgeons. Since these topics have already been described elsewhere in the text, details regarding epidemiology, diagnostic evaluation, and specific operative approaches will only briefly be discussed.


6.2 Gastrointestinal Bleeding (GIB)






  • Not a viable option for patients who present with massive gastrointestinal bleeding (GIB) and shock: operation resuscitation and localization/treatment occur simultaneously in the operating room:



    • Intraoperative endoscopic evaluation.


    • Segmental clamping of the bowel to facilitate identification of the bleeding segment.


    • Bowel resection without localization of the source is not recommended as the rebleeding rate is high (50 % for hemicolectomy when the source is not localized).


  • Indications for surgery in GIB include:



    • Failure of nonsurgical hemorrhage control


    • Ongoing hemodynamic instability


    • Transfusion of >6 units of blood


    • Difficulty cross-matching blood (due to antibodies)


    • Suspected or known malignancy (particularly with gastric ulcer)


    • Pathology not correctable without surgery (aortoenteric fistula)


  • After successful resuscitation outside the operating room, the nonoperative approach (diagnosis and potentially therapy) is by endoscopy or interventional radiology:



    • Success rates vary depending upon the etiology of the bleed and the modality chosen, but even if control of hemorrhage is achieved initially by nonsurgical means, operation may still be necessary.


    • Decision-making is complex and requires an understanding of the perceived risk of rebleeding, the underlying pathology, the morbidity associated with surgery, and the morbidity/mortality associated with failure of observation.


    • Well-documented risk factors associated with poor outcomes include age >60 years, presence of comorbid disease, shock on presentation, onset during hospitalization, persistent or recurrent hemorrhage, and need for emergent surgery.


  • Overall, 80 % of acute GIB is from an upper gastrointestinal source and is best discussed in terms of variceal and nonvariceal causes:



    • Nonvariceal upper GIB:



      • Peptic ulcer disease is the most common cause:



        • Hemorrhage is controlled in 80 % of patients following initial endoscopic intervention and 75 % of patients following repeat endoscopy.


        • Angioembolization is less effective (65 % success rate).


        • Rebleeding is associated with increased mortality and about 10 % of patients will require operation.



          • The Forrest classification is a well-described risk assessment for rebleeding based upon ulcer characteristics.



            • The presence of active arterial hemorrhage (Forrest Ia) or large, nonbleeding visible vessel (Forrest IIa) is associated with a substantial rebleed risk.


            • Ulcers >2 cm, posterior duodenal ulcers, and gastric ulcers also have a high risk of rebleeding.


        • If a patient has stopped bleeding, has numerous high-risk factors for rebleeding, and is not a prohibitive operative risk, surgery is recommended in a controlled, planned setting to avoid the morbidity of emergent surgery.


      • Other causes are less likely to require operation:



        • Mallory-Weiss tears are self-limited 90 % of the time, but if intervention is required, endoscopy is highly successful.


        • Stress gastritis is uncommon in the era of acid-suppression therapy and typically is successfully managed medically.


        • Esophagitis is generally managed medically with a high rate of success and endoscopy is useful for refractory cases.


        • Dieulafoy’s lesion is successfully treated endoscopically in 80–100 % of cases.


        • Bleeding into the bile duct or pancreatic duct (hemobilia or hemosuccus pancreaticus) is generally managed with angiography and intervention with high success rates.


    • Varicealrelated upper GIB:



      • Rarely requires operation.


      • Endoscopy is 90 % effective for esophageal varices (although repeat may be required) but is not as effective for gastric varices.


      • For the 10 % of patients who continue to bleed or rebleed, transjugular intrahepatic portosystemic shunting (TIPS) is 95 % effective in controlling bleeding.


      • Urgent surgical shunts are rarely required but can be considered in patients who have good hepatic reserve and are not transplant candidates.


    • Lower GIB (LGIB):



      • Colonoscopy is effective in identification of the source in 95 % of cases and has a low (0.5 %) complication rate.


      • Diverticular disease is the most common source.



        • Massive lower GIB originates in the right colon in two-thirds of cases.


        • Therapeutic colonoscopy is generally effective at stopping the bleed acutely.


        • If this fails or the patient rebleeds, angioembolization can be considered (success rate: 40–85 % of cases, but the rebleeding risk is high, particularly if the small bowel or the cecum is the source).


        • The overall risk of rebleeding at 1 year is 10 % but rises to 50 % at 10 years.


        • If the diseased segment has been localized, elective colonic resection is indicated for good surgical candidates.


      • Angiodysplasia can be diagnosed and treated successfully in most patients with colonoscopy or angioembolization:



        • Segmental colectomy should be performed if the lesion has been localized but continues to bleed.


        • Hemicolectomy without specific localization of the source should be avoided because of the high risk of failure to resect the pathology, with high incidence of rebleed.


        • However, if the source is felt to be the colon, or if all other causes have been eliminated, subtotal colectomy is recommended although the mortality is high (30 %) when performed emergently.


      • Tumor:



        • Usually surgery, ideally nonemergent, is required.


    • All patients who undergo angioembolization should be followed closely for signs of mesenteric ischemia (particularly patients with significant vascular disease), but the overall risk appears to be low.


  • Up to one-third of patients with LGIB actually have a small bowel source:



    • If the patient’s clinical status permits, a thorough search for the source should be performed before operation is considered.

Oct 16, 2016 | Posted by in EMERGENCY MEDICINE | Comments Off on When to Operate After Failed Nonoperative Management

Full access? Get Clinical Tree

Get Clinical Tree app for offline access