Last month, we presented a malpractice case to our readers and asked for your verdict. Here’s our final analysis.
Brady Pregerson, MD: My opinion is that the care in this case, while not ideal, did represent reasonable practice, especially for an emergency physician working in a busy urban ED. The patient presented with vomiting and abdominal pain and on exam was a bit tachycardic and was tender on exam. Appropriate initial testing was ordered but there was a delay in ordering initial IV fluids. When labs came back abnormal a CT of the abdomen was ordered but the patient waited many hours for the scan and results, during which time she had progressive decompensation. After CT results were reported, more aggressive care was started but her pulse and blood pressure continued to worsen despite a 2 liter IV fluid bolus. She eventually died before surgery could be started. Had testing and care been more aggressive from the start, perhaps she could have been saved.
Many patients present to the ED with vomiting and abdominal pain, but only a small minority will have a condition that causes such rapid deterioration as that seen in this case, and you cannot necessarily tel who they are on arrival. Patients at the onset of a self-limited viral illness or with biliary colic or gastritis or pancreatitis or even with a low-grade bowel obstruction may cal 911 and/or have similar initial vital signs to this patient. They do not all need advanced imaging, and in a busy urban ED with a queue for CT, those who may need imaging cannot all be pushed to the front of the line. There is no “code abdomen” like there is for stroke or STEMI, nor should there be. Neither should emergency physicians (EPs) be expected to routinely consult a surgeon soon after a patient with a possible surgical emergency presents to the ED. We al know surgeons. The way they usually respond to an early consult is with, “Call me when the CT is done.” The delays in care in this case, while they may appear problematic in retrospect, are actually quite routine in a busy ED, and therefore represent reasonable practice.
That does not mean we cannot learn to give better care from salient points in this case. Reasonable practice is not always ideal care, and good if not great care is what we should al be striving for. A seasoned provider knows that when abdominal pain is intermittent, generalized, has an onset after vomiting or resolves with IV fluids, anticholinergics and/or antacids, that a surgical or serious medical condition is much less likely. Conversely, when pain is one sided, constant, starts prior to vomiting or requires opiates to be relieved, a surgical condition is a more distinct possibility and advanced imaging may be required. But who needs to be moved to the front of the CT queue or have the surgeon paged before imaging is completed? To answer that question, physician judgement is required and a tallying of the number of red flags is helpful. For bowel obstruction, red flags for strangulation include the following: fever, constant pain especially if >12 hours, hypotension and/or tachycardia especially if unresponsive to fluids, hernia, prior gastric bypass surgery, abdominal rigidity, acidosis, leukocytosis especially if the WBC count is >20 or there is bandemia, and high-grade obstruction on imaging . The more red flags that are present, the more urgent should be the medical care, including fluid resuscitation, early imaging with consideration for plain films as well, and early consultation by a surgeon.
We all know that vomiting and diffuse abdominal pain without diarrhea can have many causes, but in a patient with prior abdominal surgery, bowel obstruction needs to start at or near the top of the list. Documentation in this case implies pain out of proportion to physical exam, which is a concerning finding for ischemic bowel, a known and potentially fatal complication of bowel obstruction. In addition, initial vital signs and lab values in this case represent multiple red flags making evaluation and management of this patient potentially emergent. Initial orders should have probably included IV fluids. If the patient was not improving, especially given the history of prior gastric bypass putting the patient at risk for an internal hernia, a type of closed-loop bowel obstruction, then imaging and consultations should ideally have been expedited. Options would have included starting with plain films or pushing the patient to the front of the CT queue and not adding delays, such as those associated with oral contrast.
Oral contrast provides limited to no utility in evaluating for most of the emergent causes of acute abdominal pain and can lead to significant delays in imaging when protocols require certain durations of contrast dwell time or when patients cannot finish drinking the contrast due to nausea, vomiting or pain. In addition, though IV contrast can help detect bowel ischemia when the bowel wall fails to enhance, oral contrast can actually hinder this detection . The use of oral contrast in abdominal CT’s should therefore be on a selective basis, if at all. I still remember one patient who was signed out to me years ago who was waited hours to get her CT with oral contrast, only to die in the ED just before they were ready to take her to surgery for a high grade bowel obstruction. Ironically, gastric bypass patients may benefit by a special protocol with a small volume of oral contrast and a very brief dwell time.
It is important to know that closed-loop bowel obstructions are more likely to cause strangulation, necrosis and perforation and therefore usually require emergency intervention. Causes of closed-loop obstructions include hernias, including internal hernias, and volvuli. A gastric bypass patient who has lost significant weight is at risk for an internal hernia due to their altered abdominal anatomy. This can be difficult to diagnose clinically as there may be no vomiting, and can even be missed on CT scan. As it can be fatal if not expeditiously treated, one should have a low threshold for surgical consultation in any gastric bypass patient with unexplained abdominal pain and/or vomiting.
In summary, this patient died from a rapidly progressive, closed-loop, small bowel obstruction caused by an internal hernia. Her prior gastric bypass surgery but her at risk for this condition. Although not ideal, the care she received in the ED was reasonable, with delays that are not atypical in a busy urban ED that may have multiple simultaneous patients with abdominal pain and vomiting. She had a thorough evaluation including blood work, advanced diagnostic imaging and surgical consultation. She was treated with appropriate analgesia and rehydration. Unfortunately, she had a deadly condition that was not immediately apparent. Ideal care might have included earlier recognition of her unique risk factor for this rapidly progressive type of bowel obstruction, earlier identification of the red flags in her case and expediting earlier aggressive resuscitation, and earlier diagnostic testing and surgical consultation. Though one cannot be certain, perhaps with that she might have had a chance. As emergency physicians though we should be judged on what is reasonable, we should nevertheless strive to be exceptional.
William Sullivan, DO, JD: Difficult case. A 30-year-old patient presented with acute onset of abdominal pain and non-bloody emesis. She had diffuse abdominal tenderness, but no rebound or guarding. Most emergency physicians probably see patients with similar presentations each shift. But this patient also had some reasons for concern. A WBC of 16 isn’t particularly impressive, but a bandemia of 18% certainly grabs one’s attention. Her pain was out of proportion to her physical examination findings. The patient also had a history of abdominal surgery – and not just any abdominal surgery, but a gastric bypass.
The management of abdominal pain isn’t a one-size-fits-all approach. Brady provided a nice summary of general considerations in patients presenting with abdominal pain. However, patients with abdominal pain who have had gastric bypass surgery may deserve special consideration. A Roux-en-Y is commonly performed in patients desiring weight loss. In this procedure, the stomach is transected and a small pouch of gastric mucosa is created. A portion of the distal intestine is then anastomosed to the gastric pouch and the remainder of the stomach and proximal portion of the intestine is anastomosed to the distal GI tract. Weight loss is caused both by the restrictive size of the gastric pouch along with some degree of malabsorption due to bypass of much of the proximal GI tract. While there are several other weight loss surgical procedures currently being performed in the US, it is the Roux-en-Y procedure that was important in this case.
Post-gastric bypass patients may develop abdominal pain for many reasons. Overeating and malabsorption may lead to crampy abdominal pain and vomiting. Patients may also develop “dumping syndrome” when they ingest large amounts of sugars that bypass the proximal bowel and cause abdominal cramps and osmotic diarrhea further downstream. Gastric bypass patients with abdominal pain are also at risk for more serious problems. Post-surgical adhesions can cause pain and/or bowel obstructions. Stenosis of the surgical anastomosis site can also occur. Gastrointestinal bleeding or perforation may occur due to mechanical tension on the anastomosis site between the distal small bowel and the gastric pouch. Incisional hernias can sometimes occur and may be difficult to detect due to the patient’s body habitus. However, the patient in this case suffered from an uncommon but serious complication of gastric bypass surgery – a Petersen’s hernia. Petersen’s hernia occurs when a loop of bowel is trapped in the opening between the alimentary loop of small bowel (distal bowel anastomosed to the gastric pouch) and the colonic mesentery. The incidence of a Petersen’s hernia after a Roux-en-Y procedure ranges from 0.9% to 4.5%. Patients who have lost a large amount of weight are at higher risk for this type of internal hernia because spaces that had been sewn closed may pull open as the intra-abdominal fat contracts. The problem with diagnosing Petersen’s hernia in gastric bypass patients is that the presentation can be variable – anywhere from vague intermittent pain to constant severe pain with bowel obstruction and necrosis as was seen in this patient. Diagnosis of many of the post-gastric bypass complications requires a CT scan of the abdomen, preferably with oral and IV contrast, keeping in mind that with a small gastric pouch, the patients will not be able to tolerate the same volume of oral contrast as patients with normal abdominal anatomy.
Did the physician’s management of this patient represent reasonable practice? The plaintiff expert criticized the physician’s failure to order a CT scan and consult a surgeon on arrival. Knowing the patient’s outcome, these criticisms may seem reasonable, but unfortunately, they appear to be influenced by retrospective bias. It simply isn’t a reasonable practice to order immediate CT scans and urgent surgical consults on every 30 year old patient presenting with acute abdominal pain, vomiting, and no rebound tenderness. Brady noted that this patient may have survived if testing and treatment had been more aggressive from the time the patient arrived in the emergency department. I’m not so sure. Suppose the patient received an additional liter IV fluid bolus upon arrival. I think it is still unlikely that the extra fluid would have prevented or delayed the patient’s decompensation after the CT scan. On presentation to the emergency department, the patient’s abdomen was diffusely tender, but had no rebound or guarding. Even if the patient was diagnosed with a bowel obstruction at that point, it is unlikely that the surgeon would have rushed the patient to emergency surgery. Waiting to see whether the patient’s symptoms responded to treatment and waiting for return of lab results was reasonable at that point. While the CT scan demonstrated a high-grade bowel obstruction and a “possible internal hernia,” we don’t know what specific CT scan findings led to the radiology report of a “possible internal hernia” and whether those findings would have been present without oral contrast. Brady’s summary correctly notes that oral contrast with CT scanning provides limited utility in determining causes of acute abdominal pain. However, oral contrast CT scans are still useful in patients with known inflammatory bowel disease, GI tract-altering surgery, and patients with thin body habitus. In this case, an earlier CT scan may have helped reach the diagnosis of a small bowel obstruction sooner, but foregoing oral contrast to expedite the CT scan may also have made the diagnosis of an internal hernia more difficult or caused the diagnosis to be missed altogether. By the time the CT scan was completed, the patient’s physical examination had changed and her vital signs had decompensated. It was then up to the surgeon to determine when the patient was stable enough to go to surgery. She received appropriate resuscitation but unfortunately suffered from a rapid progression of an uncommon problem. Even though there were some delays in management, those delays are typical in most emergency department settings. The process of providing emergency medical care – especially in a patient with complex medical problems – necessarily involves many sequential steps, each of which takes time to complete. Rushing this process may save time in some cases, but may also cause errors in other cases. This patient’s care was reasonable and reflected the time constraints present in any busy emergency department. Even if this patient received “perfect care,” I agree with the defense expert’s opinion that her outcome would likely have been the same.
“There was a delay in surgical consultation which should have been obtained upon noting abnormal CBC ,metabolic acidosis and not noted nut likely persistent abnormal vital signs. Probably the final outcome could have been unchanged since patient died within nine hours of onset of pain.”
– Marcel Cesar
“Sounds entirely reasonable and appropriate to me. We are taught that bariatric surgery patients should always get IV and oral contrast and this takes at least 2 hours. Does not sounds like plain films would have shown anything and physician’s documented exam does not suggest acute abdomen. Calling the surgeon on arrival for every patient complaining of severe abdominal pain is laughable. Perhaps the only thing that should have been done differently was faster administration of IV fluids.”
“Not optimal, but reasonable. I know that plain film x-rays rarely change management, but in this case might have been able to provide a more rapid diagnosis of the bowel obstruction. Also, the literature shows that oral contrast is not needed for Emergency Department patients, and a non-contrast CT could have been obtained more quickly.”
“Why wait for labs to come back before ordering a CT in a case like this? How would they change management? A patient with gastric bypass history and a presentation described above is at high risk for bowel obstruction. Also, why oral contrast? According to the most recent guidelines from the radiologists, oral contrast is not required to diagnose bowel obstruction or most of other serious acute abdominal conditions. A case like this warrants early surgical involvement. A more reasonable course of action would be to order CT with iv contrast only right after evaluating the patient and giving heads up to the surgeon. Personally, I would get obstructive series while resuscitating/waiting for more definitive test to get done to be able to make a stronger case to the surgeon hoping that he/she might be convinced to be more aggressive and take the patient to OR based on the clinical picture and obvious signs of obstruction on plain x-rays without waiting for CT to get done. A patient with the given previous surgical history and the given clinical presentation requires aggressive management. Hard to make a case here that the care was reasonable.”
– Jerry Guzik
“Step-by-step management for this case is right. But I think that an abdominal X-Ray and upright CXR make that EP a quicker diagnosis. I agree with that surgeon about prognosis. Busy ED is not a reason for delay in management for the court, unfortunately.”
– Mahmoudreza Sajjad
- Pregerson, DB, Quick Essentials Emergency Medicine 1-minute Consult, EMresource.org
- Pregerson, DB, Tarascon Emergency Department Quick Reference Guide, Jones & Bartlett