Nonoperative management is the standard of care for most solid organ injuries, including the liver. More serious injury may require operative intervention. Unlike the spleen, however, the liver has a higher complication rate when managed nonoperatively or operatively. One of the more troubling problems is the persistent bile leak. Our radiology colleagues do a great job a draining collections, but what should we do if the bile keeps pouring out?
ERCP seems like a reasonable choice. But does it work? The Shock Trauma Center looked at their experience over a 6 year period. They included both blunt and penetrating injuries to the liver, and found a total of 26 patients in their database. All but 2 underwent an initial attempt at operative control of the bile leak. All but one had ERCP performed within 3 weeks of admission.
They found that ERCP resulted in decreased drain output within 2 days. All bile leaks stopped within 7 months, with an average closure time of 47 days. There were no complications from ERCP itself.
Bottom line: consider ERCP part of your armamentarium when dealing with major liver injuries. Depending on patient condition, it might even be used as the initial approach to controlling a bile leak. If the leak does not decrease significantly or close in a reasonable period of time (not yet defined), operative intervention will still be required.
Reference: Endoscopic retrograde cholangiopancreatography is an effective treatment for bile leak after severe liver trauma. J Trauma, in press, 2011.
It’s always interesting to review the trauma literature of days gone by to see where we’ve been and how it impacts where we are today in trauma care. Here are a few articles from the Jan 1990 Journal of Trauma (Volume 30 Number 1) worth commenting on:
Efficacy of Liver Wound Healing by Secondary Intent. Dulchavsky et al, page 44-48. This paper compared wound healing using tensile strength in pigs and dogs. The authors compared primary operative closure, closure with an omental buttress, and healing by secondary intention. They found that the strength of secondary healing equaled or exceeded that in both types of operative repair by 6 weeks post-injury. This paper and several similar ones laid the groundwork for our understanding of solid organ healing and lend weight to the somewhat arbitrary guidelines of resuming full physical activity after 6 weeks.
Intestinal Injuries Missed by Computed Tomography. Sherck et al, page 1-7. The authors retrospectively looked at 10 CT scans done over a 9 year period that were done in patients who eventually were found to have an intestinal injury. The injury became apparent in 2 hours to 3 days after the traumatic event. Even when the authors knew that a bowel injury was present, they could definitively diagnose the problem on the initial CT in only 2. The authors concluded that CT could not reliably detect these injuries. Little has changed since this paper was published, even though the scan technology has improved greatly (1 or 2 slice scanners in 1990, 16-64 slices now). We have gotten better at detecting bowel injury with better resolutions, but the diagnosis still remains a clinical one.
Techniques of Splenic Preservation Using Fibrin Glue. Shoemaker et al, page 97-101. The senior author first described the use of fibrin glue in splenic injury in 1983, and continued to investigate it over the next 7 years. This paper was the largest human series at the time. The authors found that it limited blood loss and transfusions, although there was no actual control group. They found that it increased splenic salvage rates to 86% in operative cases, and repeat CT did not show rebleeding or abscess formation. This study added a new technique to the trauma surgeon’s armamentarium in dealing with solid organ injury. Although later studies did find a modest increase in abscess formation, the technique remains a viable alternative when operatively managing solid organ injury. Overall, it is not used as much now because nonoperative management has become quite refined, with a success rate of about 93%.