An Anesthesiologists Focus On Sedation And Complications Of GI Endoscopy Cases- A Case Of Bilateral Pneumothoraces And Pneumomediastinum After Routine ERCP
ercp, pneumomediastinum, pneumothorax, sedation
U Galway. An Anesthesiologists Focus On Sedation And Complications Of GI Endoscopy Cases- A Case Of Bilateral Pneumothoraces And Pneumomediastinum After Routine ERCP. The Internet Journal of Anesthesiology. 2009 Volume 23 Number 2.
The majority of patients require some form of sedation for gastrointestinal procedures. Traditionally sedation has been administered by endoscopists. Recently, in some U.S States, there have been changes in the rules regarding nurse administered deep sedation. That, combined with an increase in the number and complexity of endoscopy cases and endoscopies being performed on higher risk patients, anesthesiologists are being asked more and more to be involved in these cases to administer deeper and more sophisticated forms of sedation or in many cases general anesthesia. Because sedation for these endoscopies was previously performed by the endoscopists, we may imagine that these are noninvasive, low risk and safe procedures. This may lull us into a false sense of security. Furthermore, the environment and procedures of the GI endoscopy suite may be unfamiliar to the anesthesiologist performing the sedation. As a result we need to be educated and vigilant to the potential complications that may occur during routine endoscopies.
Our case is one in which the major complication of duodenal perforation, bilateral pneumothoraces and pneumomediastinum occurred during ERCP with anesthesia administered deep sedation.
Our patient was a 56 y/o female who presented for an elective outpatient endoscopic retrograde cholangiopancreatography (ERCP) to investigate abdominal pain and cramping which had been present for 8 months. An outside hospital CT scan noted intra and extrahepatic biliary dilation. Past medical history included hypertension, COPD, mild GERD and hyperlipidemia. Past surgical history included open cholecystectomy. She was initially evaluated in the preoperative area by the anesthesia team. MAC (monitored anesthesia care) sedation administered by anesthesia providers was discussed with her. She was then brought to the ERCP suite by the anesthesia team and placed in the semi prone position. Full ASA monitors were applied as well as oxygen via nasal cannula. The patient was sedated with an infusion of propofol 9 mg/cc combined with alfentanil 45mcgs per cc. Once the patient reached a deep level of sedation a nasal trumpet was inserted to alleviate mild upper airway obstruction and the endoscope was introduced and advanced. She was maintained on a propofol/alfentanil continuous infusion. Vital signs remained stable and the patient maintained an oxygen saturation of 97% on 4 liters nasal cannula while breathing spontaneously throughout the procedure.
A periampullary diverticulum was found with the ampulla inside it. The lip of the diverticulum had to be lifted with the sphinctertome in order to see the ampulla. The bile duct could not be selectively cannulated. The pancreatic duct was cannualted and a guide wire was placed inside the pancreatic duct without injecting contrast. A pancreatic sphincterotomy was then performed. Several attempts were made at cannulaion of the bile duct however it proved to be very difficult. While attempting cannulation of the bile duct, the patient desaturated to 80%, other vital signs were stable. She was though to be obstructing secondary to the sedation so the decision was made to abort the procedure, discontinue the sedation and turn the patient to the supine position. The patient was placed supine and in the sitting position, 100% oxygen via a non rebreathing mask was administered. She was breathing spontaneously, responding normally to commands and alert and orientated times three. Her vital signs were normal except for a saturation of 86% on 100% oxygen via a non rebreathing mask. Upon examination, we found her to have excessive subcutaneous crepitations on her right and left chest wall, extending to her neck. The patient was emergently transferred, fully monitored to PACU for further evaluation.
A chest x ray was ordered which showed moderate right pneumothorax associated with extensive subcutaneous emphysema, free intraperitoneal air and suspected pneumomediastinum (figure1). A right sided chest tube was placed. Her vital signs remained stable and she did not require intubation. A CT of the chest and abdomen were then performed which showed extensive mediastinal free air, involving all compartments of the mediastinum from the thoracic inlet to the diaphragm, a large amount of subcutaneous emphysema and small bilateral pneumothoraces (figure2). The abdominal CT showed large amounts of retroperitoneal air, moderate amount of free intraperitoneal air and subcutaneous emphysema along the right flank.
The patient was brought to the operating room for laparotomy and exploration after abdominal distention began to worsen on post procedure day number 1. She was found to have a duodenotomy in the second portion of her duodenum and extensive free bile in the peritoneal cavity. She had a resection of her duodenal diverticulum and over-sewing of her duodenotomy. She was transferred to the surgical ICU postoperatively.
Her postoperative course was complicated by an enterocutaneous fistula through her right upper quadrant incision which was treated with catheter drainage and a wound VAC. She also developed two subsequent retroperitoneal abscess which required operative drainage. She had a prolonged hospital course but was eventually discharged 24 days after her initial ERCP.
Here we described an elective ERCP done under anesthesia administered sedation which had several severe adverse outcomes. Anesthesiologists often think of endoscopy procedures as benign procedures, given that the endoscopists themselves often both simultaneously perform the procedure and administer the sedation. However we must be vigilant to the adverse outcomes that can occur from a seemingly benign procedure.
When exactly is an anesthesiologist needed for GI procedures?
First let us define moderate (conscious) versus deep sedation.
Moderate (Conscious) Sedation
Moderate sedation (conscious sedation) is a drug-induced depression of consciousness during which patients respond purposefully to verbal commands, either alone or accompanied by light tactile stimulation. No interventions are required to maintain a patent airway, and spontaneous ventilation is adequate. Cardiovascular function is usually maintained. 1
Many endoscopy procedures are done under moderate sedation. Examples of procedures that require moderate sedation are EGD and Colonoscopies. Moderate sedation is usually administered by the endoscopist with the assistance of trained nursing personnel. Generally they use a combination of benzodiazepine and opioid such as midazolam with morphine, meperidine or fentanyl.
Deep sedation is a drug-induced depression of consciousness during which patients cannot be easily aroused but respond purposefully following repeated or painful stimulation. The ability to independently maintain ventilatory function may be impaired. Patients may require assistance in maintaining a patent airway, and spontaneous ventilation may be inadequate. Cardiovascular function is usually maintained. 1
Deep sedation is generally required for a number of complex procedures that are more invasive, usually involving stenting, biopsy or drainage. Examples of procedures requiring deep sedation include:
DOUBLE BALLOON ENDOSCOPY
FEEDING TUBE PLACEMENT
So when is an anesthesiologist needed for administration of sedation for endoscopy? There are several issues at hand which have recently changed the role of the anesthesiologist in the GI suite. In 2007 the Ohio Board of nursing changed its laws regarding nurse administered sedation. Here in Ohio nurses can now only administer moderate (conscious) sedation. Nurses can no longer administer deep sedation. Cases requiring deep sedation require the presence of an anesthesiologist to administer the deep sedation. The Nursing Interpretive Guidelines now state that “The nurse should not engage in activities that are the practice of anesthesia care…Therefore should not administer medications to induce deep sedation and/or anesthesia” 2
There are several additional situations where the endoscopist may request an anesthesiologist for the administration of even moderate sedation. Examples include:
History of difficult sedation in the past
History of problems with anesthesia
Delirium or dementia
Chronic pain patients
Significant medical co-morbidities
Another reason for the increasing need to have an anesthesiologist present to perform deep sedation, is the ever increasing complexity of the endoscopy cases. Procedures have become more complex and instruments sophisticated. This leads to the need for more effective sedation and analgesia techniques and the need to leave the endoscopist free to concentrate on the procedure at hand. 3
Complications Of ERCP
ERCP is performed as a diagnostic and therapeutic tool for a variety of biliary and pancreatic problems. Risk factors for complications include sphincter of oddi dysfunction, difficult cannulation, precut sphincterotomy in inexperienced hands, failure to achieve drainage and percutaneous transhepatic biliary access. 4
Complications of ERCP include
Hemorrhage (usually after sphincterotomy) 1.34%
Perforation of viscus 0.6%
Overall mortality is 0.33%. 5
Our patient experienced the complication of duodenal perforation and subsequent pneumomediastinum and bilateral pneumothoraces.
Perforation during endoscopic sphincterotomy is a well recognized complication. It is usually retroperitoneal. 6 Retroperitoneal perforation occurs due to extension of a sphincteroromy beyond the intramural portion of the biliary or pancreatic duct. Intraperitoneal perforation may occur due to perforation of the bowel wall by the endoscope. Our patient had both intraperitoneal and retroperitomeal air from a duodenal perforation. Risk of perforation is less than 1% of ERCPs and sphincterotomys. Perforation is more likely in patients with Billroth 2 anatomy, needle-knife sphincterotomy, and sphincter of oddi dysfunction. 4 The presence of juxtapapillary diverticulum has not been shown to increase risk of perforation. 6, 7, 8
Pneumothorax and pneumomediastinum resulting from perforation after ERCP and sphincterotomy has been reported many times in the literature. 9,10,11,12,13,14,15,16,17 The extensive emphysema is thought to be due to excessive air insufflation during the procedure after the perforation has occurred. The retroperitoneum is divided by the anterior and posterior renal fascia into three compartments which are in communication. After duodenal perforation air enters the right anterior pararenal space and then can further travel inferiorly and communicate with the posterior pararenal compartment. The posterior pararenal compartment extends through the diaphragmatic hiatus and into the mediastinum. Thus air under pressure enters the retroperitoneum and travels into the mediastinum causing pneumothoraces and pneumomediastinum as well as subcutaneous emphysema. 9 The amount of air leakage may not reflect the size of the perforation as much as the extent and duration of manipulation performed after the perforation. 6
Duodenal perforation is best diagnosed by abdominal CT.18 The treatment of perforation depends on the clinical state of the patient. Conservative management is recommended for patients with minimal signs, normal white cell count and absence of fever. This involves treatment with antibiotics, nasogsatric suction and nothing by mouth. Surgical management with closure of the perforation and drainage of the retropetioneal space is recommended for patients with signs of sepsis, contrast leakage or evidence of collections. 19
Due to recent changes in nurse administered sedation, coupled with the need for more sophisticated forms of sedation for the increasing number and complexity of endoscopies being performed, anesthesiologists are now more commonly seen in the GI suite administering sedation as well as general anesthesia. As out report here shows, not all GI endoscopy procedures are free from major complications. Therefore we must be vigilant to these possible disastrous events as we anesthetize and monitor these patients.