Dodging the Bullet

A review and discussion of several close clinical encounters

By Larry Mellick, MD, MS, FAAP, FACEP, Editor-in-Chief, Professor of Emergency Medicine and Pediatrics, Department of Emergency Medicine, Medical College of Georgia, Augusta

In this article, we present a series of actual clinical scenarios that could have turned out differently if the wrong management decision had been made. There are two goals of this article. The first is to glean from each of the reported cases important points of educational value and learning. The second is to point out that clinical misadventures are often a single judgment call away from a potential tragedy. Every day, emergency medicine physicians find themselves in similar situations. Unfortunately, some disease presentations are not classic textbook descriptions, and atypical clinical presentations are common. Besides complex and atypical presentations, the emergency department is a relatively uncontrolled environment. We work under conditions that are cognitively demanding. Emergency health care providers experience frequent interruptions, care for a wide range of patients simultaneously, and experience surges in multiple patient care responsibilities.1,2,3 The decision-making processes can be complicated, tenuous, and treacherous. The odds are not necessarily in our favor. Consequently, for every bullet successfully dodged, we are reminded that it may be just a matter of time before one finds its mark. If and when that happens, we can only hope that harm is minimized, our documentation supports our decision-making, and that the patient and his or her family are understanding and forgiving.

Blood Loss and the Pediatric Patient

A 6-year-old girl who lived in a rural area was playing outside with a stray dog recently befriended by the family. When the child's mother heard her screaming, she ran out of the house to find her daughter covered in blood and the dog standing quietly at her side. The little girl had sustained a major scalping laceration with a large anterior flap that exposed the skull bone. Emergency medical services responded to the emergency call and, at the mother's insistence, a decision was made to transport the young girl to the children's hospital emergency department rather than their local emergency department. The prehospital providers described what appeared to be a large amount of blood loss at the scene. The gauze that covered the huge scalp wound was soaked with blood, and the wound continued to ooze blood from small arterial bleeders. (See Figure 1.)

Figure 1: Blood, blood-soaked gauze, and elastic wraps

The initial vital signs were blood pressure 103/57, heart rate 142, and the pulse oximeter reading was 100%. One 20-gauge intravenous catheter had been successfully placed in the right wrist. Despite multiple attempts, a second intravenous line could not be obtained. Pulse pressure, heart rate, capillary refill, and mental status were the parameters available for assessment of the patient's hemodynamic stability. Early in the patient's care, fentanyl was suggested for pain control. The patient was already quiet and, while responsive to the pain from attempted needle sticks, already seemed a little too sedated. Since mental status was one of the few parameters initially available for monitoring the patient's circulatory status, we elected not to treat with fentanyl immediately. The normal pulse pressure is 30 to 40 and narrows with increased systemic vascular resistance. Despite the large amount of blood lost, her pulse pressure had not narrowed. Since bone injury and skull penetration are not uncommon following dog bites (which can exert a force more than 400 pounds per square inch), a computerized tomography (CT) scan of the head was obtained. The CT scan of head was normal.

The plastic surgery team was consulted, and they evaluated the patient in the emergency department. Because the child had something to drink at 10:00 a.m., the surgeons did not want to take the child to the operating room until 4 p.m. Furthermore, the consultants asked that the planned blood transfusion be held. Meanwhile, the ongoing blood loss from the scalp wound was proving resistant to our attempts at hemostasis. Despite pressure dressings, various hemostatic agents, and strategically placed staples, the oozing continued. The ongoing bleeding hidden under large amounts of gauze and pressure dressings was almost clandestine. Only when the dressings were removed to assess hemostasis did the continued blood loss become apparent. A repeat hemoglobin measurement demonstrated a dramatic drop from 12.0 to 8.2 g/dL. Because of the drop in hemoglobin and the planned delay in surgery, a transfusion of packed red blood cells was finally started. Later that afternoon, the patient's large scalp wound was repaired in the operating room.


Blood loss in children can be easily underestimated. Furthermore, the total blood volume of a child is considerably less than that of an adult. Blood loss that might appear small by adult standards may very well be life-threatening for a child. And, as was described in this case, blood loss at the scene that appears large is more likely a critical loss of blood for a child. Additionally, despite massive blood loss, physiologic adjustments in systemic vascular resistance occur that maintain the child's vital signs at a near normal state. Once these compensatory mechanisms fail, sudden deterioration and death can occur. Our patient initially demonstrated a normal capillary refill, normal pulse pressure, and a normal skin color. Urine output was not immediately measured because urine standing in the bladder is of no clinical value. Only after a Foley catheter is placed, the bladder is emptied, and tracking is started is urine output of potential value. The only notable parameter on this child was the change in mental status. Nevertheless, the child demonstrated laboratory evidence of a major, life-threatening blood loss.

Delaying emergency, and possibly life-saving, care because a patient recently ate or drank is not well supported in the literature. At least for procedural sedation, there is sufficient evidence and expert consensus that the risk is small and the management of the emergency condition can take priority.4,5,6 Even guidelines published by the American Society of Anesthesiologists indicate that in emergency situations, recent oral intake is not a sufficient reason to delay surgery, but the target level of sedation can be modified.7

Learning Points

• Never completely trust your own estimations of pediatric blood loss.

• Respect the pediatric patient's ability to maintain normal vital signs despite major blood loss.

• Continually track and reassess the bleeding pediatric patient's physiologic status.

• Err on the side of early fluid and blood replacement.

• Be very mistrustful (even paranoid) of bleeding scalp lacerations.

• When a child last ate or drank should be a minimal consideration when it comes to timely management of his or her emergency condition.

A Fish Bone Story

A 17-year-old girl presented to the emergency department complaining that she had a fish bone stuck in her throat. She described an area of irritation in her right posterior pharynx. The initial examination with a tongue blade demonstrated no evidence of a fish bone. The entire area was viewed with a nasopharyngeal scope, and no evidence of a foreign body was noted at her vocal cords or valleculae. A radiograph of her neck demonstrated no foreign body. The search for a foreign body was nearly ending when the patient pointed out a more specific area of concern in her tonsil. (See Figure 2.) Further examination demonstrated the nubbin of an object that was slightly protruding from the tonsil. Using forceps from a suture kit, the emergency medicine resident extracted a fish bone that had been impaled into the tonsil. (See Figure 3.)

Figure 2: Small area of irritation in the patient's right pharynx

Figure 3: Fish bone removed from the teenager's right tonsil


Emergency physicians use a number of tools to search for pharyngeal foreign bodies. But what happens when these investigations fail to demonstrate a fish bone? Usually, the assumption is made that the persisting pain is secondary to an abrasion caused by a sharp foreign body passing through the pharynx. This case demonstrates that an impaled foreign body no longer visible can also be the cause of irritation. There are numerous case reports that describe the consequences of chronically impaled fish bones, and confirm that the associated morbidity to the patient can be severe.8,9,10

Learning Point

• If one cannot actually see the fish bone, it may be impaled into the tissues. Be careful about attributing persistent pain to a scratched pharynx.

Chest Contusion Conundrum

A 52-year-old man presented to the emergency department complaining of right-sided chest pain and "bruised ribs," as well as recent hemoptysis. (See Figure 4.) During the previous week, he lost his balance and fell against a coffee table. Other than some chest discomfort with inspiration, he denied any other complaints. In fact, most of his discomfort occurred when he was lying flat in bed. His past medical history was significant for diabetes mellitus, hypertension, and a lower extremity amputation below the knee several months prior. He denied any lower extremity swelling or pain. His initial vital signs were pulse oximetry 99%, blood pressure 116/66, heart rate 106, and respiratory rate 16. Subsequent vital signs showed a heart rate of 93, 101, and 97. Pulse oximetry readings remained at 99% and 97%. His chest radiograph demonstrated an area of haziness and a plate-like atelectasis that was considered consistent with an infiltrate or even a pulmonary contusion. (See Figure 4.)

Figure 4: Abnormal chest radiograph of the patient

The possibility of a pulmonary embolus was considered. However, the radiology findings were not classic ones for pulmonary embolus. Additionally, on repeated questioning, the patient denied any shortness of breath and stated that he "felt fine." Preparations were begun to discharge the patient home with a prescription for azithromycin to treat presumptive pneumonia. At the last minute, however, uneasiness with the decision resulted in a physician ordering a CT angiogram (CTA) of the chest. Besides emphysematous changes in the lungs and peripheral airspace opacities in the bilateral lower lobes, the CTA demonstrated bilateral pulmonary emboli with significant clot burden and deep vein thrombosis in the bilateral lower extremities.


This patient did have several clinical findings consistent with a pulmonary embolus. The history of recent surgery was relevant, as was the pleuritic chest pain and hemoptysis. However, these same signs and symptoms could also have been consistent with a diagnosis of pulmonary contusion or pneumonia. The confounding historical variables that almost caused an errant clinical decision were the history of a recent fall and chest wall contusion. The lack of bruising to the chest seemed inconsistent with a significant injury and was one factor that caused the diagnosis to be reconsidered. The normal vital signs (except for the initial heart rate that demonstrated a tachycardia) and the pulse oximetry of 99% were inconsistent with the diagnosis of a pulmonary embolus, as was the patient's lack of symptoms. The patient repeatedly denied shortness of breath and stated that he "felt fine." The radiograph was abnormal but did not demonstrate the classic pulmonary embolus radiograph findings of Westermark's sign or Hampton's hump.

Learning Points

• Pulmonary emboli presentations can be subtle, and vital signs (heart rate, pulse oximetry, etc.) are not reliable indicators.

• Always beware of the "red herring" in the history that can distract you from the true diagnosis.

• Be sure that you have sufficiently ruled out the worst possible scenario before accepting your final diagnosis.

Last Rites

An elderly woman in her late 80s was transferred from an outside hospital for management of a subarachnoid hemorrhage, as well as large amounts of free air discovered in her abdomen. After the neurosurgery team placed a pressure monitoring device in her head, the gastrointestinal surgery service was preparing to take her to the operating room. However, before any additional steps could be taken, the patient's son called and stated that the family wanted to withdraw care. The son was asked to fax a copy of the advanced directive to the emergency department. According to the advanced directive, the son clearly had decision-making responsibilities for his mother, but the patient had also clearly indicated that she wanted everything possible done to keep her alive. When the son called back on a recorded line to restate his wishes, the contradiction between his request and his mother's explicit wishes was discussed. The patient's son acknowledged the disparity between his request and the mother's wishes. And, in the same conversation, declined to come and be at his mother's bedside when she died. "We said our goodbyes last night just before she was transported to MCG." He said.

Figure 5: Priest, chaplain, and nurse perfoming last rites on the patient

The obvious dilemma required careful management. Risk management was consulted. Members of the risk management department also communicated with the son. Additionally, an ethics committee meeting was set up for 1 p.m. the same day. The neurosurgeons and the gastrointestinal surgeons were present for this meeting. The consensus was that the patient's condition was indeed futile, and a decision was made to withdraw care, as requested by the son. Because the family was of the Catholic faith, a request had been made by the son for last rites to be performed. Last rites were performed earlier in the day by a priest and the hospital chaplain. (See Figure 5.) The patient was given morphine for comfort, and the endotracheal tube was removed. The ventilator was turned off, and the dying process commenced. Over a period of several hours, the patient's blood pressure and oxygenation continued to drop. Just after 5 p.m., the patient went into an agonal rhythm and was pronounced dead.


For the emergency medicine physician, similar situations are not uncommon. Despite apparent futility, the potential for violating the wishes and rights of a dying patient exists, and decision-making must be carefully navigated.11,12,13 The AMA Council on Ethical and Judicial Affairs recommends a process-based approach to addressing futility, to include such actions as the following:14

1. Careful deliberation and resolution over what constitutes futile care;

2. Joint decision-making with physician and patient or proxy;

3. Assistance of a consultant or patient representative; and

4. Use of an institutional committee (i.e., ethics committee).

Learning Point

• When ethical dilemmas develop during the care of a patient, don't attempt to make these decisions alone. Bring in your clinical consultants, risk management representatives and, if necessary, convene an institutional committee to carefully deliberate the issues.


During every clinical shift, emergency medicine physicians find themselves making clinical decisions that have potential life or death consequences. Other decisions have the potential for serious patient morbidity. If an error in judgment occurs, harm may come to the patient, and the health care provider is at risk for allegations of malpractice. Unfortunately, there are many clinical situations that are not clear cut or in which the patient's own historical contributions are flawed. In these situations, the emergency medical care provider must rely on situational awareness, cognitive dissonance, instinct, and experience to help prevent clinical misadventures. Ordering additional studies or seeking input from colleagues or consultants may be the best decisions when significant uneasiness remains about one's clinical decision.


1. Chisholm CD, Weaver CS, Whenmouth L, Giles B. A task analysis of emergency physician activities in academic and community settings. Ann Emerg Med. 2011;58(2):117-122.

2. Chisholm CD, Dornfeld AM, Nelson DR, et al. Work interrupted: A comparison of workplace interruptions in emergency departments and primary care offices. Ann Emerg Med. 2001;38(2):146-151.

3. Chisholm CD, Collison EK, Nelson DR, et al. Emergency department workplace interruptions: Are emergency physicians "interrupt-driven" and "multitasking"? Acad Emerg Med. 2000;7(11):1239-1243.

4. Molina JA, Lobo CA, Goh HK, et al. Review of studies and guidelines on fasting and procedural sedation at the emergency department. Int J Evid Based Health. 2010;8(2):75-78.

5. Green SM, Roback MG, Miner JR, et al. Fasting and emergency department procedural sedation and analgesia: A consensus-based clinical practice advisory. Ann Emerg Med. 2007;49(4):454-461. Epub 2006 Nov 1.

6. Mace SE, Brown LA, Francis L, et al. Clinical policy: Critical issues in the sedation of pediatric patients in the emergency department. Ann Emerg Med. 2008;51(4):378-399, 399.e1-57.

7. American Society of Anesthesiologists. Practice guidelines for sedation and analgesia by non-anesthesiologists. Anesthesiology 2002;96:1004-1017.

8. Mavili E, Oztürk M, Yücel T, et al. Tongue metastasis mimicking an abscess. Diagn Interv Radiol. 2010;16(1):27-9. Epub 2009 Oct 19.

9. Jeon SH, Han DC, Lee SG, et al. Eikenella corrodens cervical spinal epidural abscess induced by a fish bone. J Korean Med Sci. 2007;22(2):380-382.

10. Kiluchi K, Tsurumaru D, Hiraka K, et al. Unusual presentation of an esophageal foreign body granuloma caused by a fish bone: Usefulness of multidetector computed tomography. Jpn J Radiol. 2011;29(1):63-66. Epub 2011 Jan 26.

11. Basanta WE. Advance directives and life-sustaining treatment: A legal primer. Hematol Oncol North Am. 2002;16(6):1381-1396.

12. Siegel MD. End-of-life decision making in the ICU. Clin Chest Med. 2009;30(1):181-94, x.

13. Rondeau DF, Schmidt TA. Treating cancer patients who are near the end of life in the emergency department. Emerg Med Clin North Am. 2009;27(2):341-354.

14. Medical futility in end-of-life care: Report of the Council on Ethical and Judicial Affairs. JAMA. 1999;281(10):937-941.