Journal Logo

Diagnosis Deconstructed

Diagnosis Deconstructed: Where the Story Begins

Morchi, Ravi MD

doi: 10.1097/01.EEM.0000407849.47091.cf
    Image
    Image:
    Image

    The patient was pronounced at 9:20 a.m. The immediate cause of death: necrotizing fasciitis secondary to Group A Strep. Associated diagnoses: Group A Strep bacteremia and streptococcal toxic shock syndrome.

    But how did it get to this point? How did we fail? Where does the story begin? I review the medical record to find out.

    The code. At 9:04 a.m. the patient transitions from sinus bradycardia with a pulse to a wide complex rhythm without a pulse. CPR is begun, and epinephrine, vasopressin, and sodium bicarbonate are pushed with no effect. No return of spontaneous circulation. The code is called at 9:20 a.m.

    Wide, agonal rhythms are almost always associated with such great contractile deficits that no palpable pulse can be generated. A pulseless state with rapid, narrow complex electrical activity, however, would have been slightly more promising.

    Multi-organ failure. At 6 a.m. the surgical progress note reads like an obituary for organs: shock liver, consumptive coagulopathy, and anuric renal failure despite aggressive resuscitation and blood product transfusion. “Dark blood continues to discharge from her right leg surgical site.”

    Operating room, second visit. At 6 p.m. the night before, the patient is taken back to the OR. The original right lower leg surgical site is explored again, and found to require further debridement of dead subcutaneous tissue, tissue that, in the presence of necrotizing fasciitis, easily dissects off the underlying fascia with finger pressure. The lower abdominal wall proximally is explored, and found to have only subcutaneous edema without necrosis. It is assumed the lower abdomen is not involved.

    The left thigh is opened broadly and necrotic subcutaneous tissue removed. The fascia is noted to be “intact.” No further blunt dissection of the subcutaneous layer could be performed. The remaining tissue seems anchored well so it is assumed that the margins of disease have been delineated. No amputation is performed.

    The illusion that surgical debridement has caught up with advancing streptococci is not uncommon in this condition. The organisms are faster and more clever than we appreciate.

    Family discussion. Five hours prior, the grave state of the patient is conveyed to family. They want to pursue life-saving measures at any cost. The surgical service offers a second operative visit as a last effort. The family agrees.

    Surgical ICU. In the hours prior to talking with the family, the patient has progressively declined in hemodynamics and organ function. The surgeon re-evaluates the right lower leg wound, and notes that what started as bullae in the calf and anterior tibial area now displays skin changes over the knee. He feels the abdomen may also be mottled. The left extremity has splotches of erythema. He makes a bedside counter-incision in the left leg, and is concerned that necrosis may be present here as well. A gram stain is sent from this leg. He writes, “My only hope is that the gram stain from the LLE is negative, and that maybe the disease has stayed on the right.” He contemplates a second trip to the OR, and calls for a family meeting.

    From the right leg to the left thigh? These organisms are not simply advancing locally through neighboring tissue. There must be hematogenous transit. They have gone systemic.

    Pathology report. At 11 a.m. a right lower leg specimen is described as having subepidermal blisters and necrosis in the deep subcutaneous tissue consistent with necrotizing fasciitis. Gram stain shows gram-positive cocci.

    Operating room, first visit. At 4 a.m. skin incisions through the lateral aspect of the right lower leg reveal gray edematous subcutaneous tissue and gray, necrotic fluid oozing from dead fascia below. A tissue specimen is sent for pathology. The muscle appears intact.

    Proximally, the right thigh is explored, and the subcutaneous layer is noted to anchor well to the underlying fascia. It is presumed that the process has spared the upper leg. No amputation is performed.

    The illusion is that surgical debridement has caught up with advancing streptococci. This sounds familiar. The organisms, however, are devious. Even at this moment, they are already in the blood, and will soon resurface elsewhere.

    Surgical consult in the ED. Called down to evaluate a confused patient with marked systemic inflammatory response and dark bullae to her right lower leg. The concern for necrotizing soft tissue infection is high. Broader antibiotics are ordered, and the patient's family consents to debridement or amputation. Her lactate is 8.

    ED re-evaluation. Patient opens her eyes to voice, but is lethargic and unable to follow commands or answer questions. She does not interact with examiners unless you squeeze her right calf or thigh. Then she responds, moaning in pain.

    An ominous finding: skin changes below her knee, but tenderness to compression far cephalad in her proximal thigh.

    Nurse note: “New development of boils to RLE.”

    EM resident note: “Ecchymosis, erythema, warmth, extremely tender to palpation, + bullae with concern for necrotizing fasciitis very high.”

    A surgical consult is called.

    It is only at this point that clinicians turned their attention to her leg. But where was the focus before?

    The brain. Head CT performed for increasing confusion; the patient is trying to get out of bed, appears anxious, and attempts to pull out her Foley catheter. The scan was normal.

    Ceftriaxone 2 g given to cover fever and confusion and the possibility of meningoencephalitis. Meanwhile, her heart rate was rising into the 130s. Her blood pressure dips to 100/50 mm Hg, despite her myocardium's efforts. Her tachypnea rising to the mid-30s. Her lactate is 5.

    The clinicians are searching in the cranium; they plan a lumbar puncture. The real source of her confusion is brewing on the other end of her body.

    ED signout. Confused and lethargic female with right lower leg bruising after a mechanical fall. Head CT and LP pending.

    ED, bed 16. Initial EM resident note: “S/P mechanical fall 3 days ago. + head trauma. No KO. LE pain yesterday. + bruising. Bilateral LE swelling since November.”

    She is tachycardic to 108, producing a blood pressure of only 120/60 mm Hg despite a past medical history of hypertension. Her temperature, originally 99.8 F, rises to 102 F. She has bilateral basilar rales and 3+ pitting edema to both feet and lower legs.

    A dysfunctional left heart at baseline; it is already beginning to fail.

    Bruising noted over her anterior tibial area on the right lower leg. Her x-ray shows no fracture.

    Anatomically, this is the closest the original ED team would come to the diagnosis. A search for fracture, at least in their thinking, is within the patient's limb. The films show no injury, and, as in the majority of cases, no subcutaneous gas to keep the clinicians' focus on the leg.

    WBC 8.2, Na 135, glucose 246, creatinine 1.

    Even if it had been calculated, her laboratory risk in necrotizing fasciitis score is only 3. She would need a markedly elevated CRP to qualify as moderate or high risk, but it was not sent.

    And here is the problem. The risk score is heavily weighted toward CRP, but that is not a lab normally sent on a patient with confusion or heart failure or bruising to her lower leg with suspicion for fracture. A lab of primary importance, its existence is entirely dependent on the correct clinical suspicion.

    Triage. A 58-year-old woman is brought in by her son for a mechanical fall three days earlier. She has tenderness and ecchymosis to the right lower leg, and both legs appear edematous.

    Presumption and expectations because of the terms “mechanical fall” and “ecchymosis” start clinicians on the wrong path. Those expectations would frame future events.

    “We were at a party, and she tripped and fell face down. She did not pass out. Her legs were swollen before the fall,” her son said.

    Afebrile at 99.8 F and heart rate 100. She complains of “very bad pain” to her right lower leg. Received one Vicodin in the medical screening area, and shortly afterward is noted to be “mildly confused” and “drowsy.” Probably the Vicodin, they say. “She always gets confused at night.”

    So this is how it starts. Pain and tachycardia only. Not even fever. Skin findings attributed to mechanical trauma, and mild confusion blamed on one tab of Vicodin and the late hour.

    The illusion is that our ears and eyes and fingers can detect advancing streptococci. Both surgeon and emergency physician are duped. The organisms meanwhile spread rapidly along fascial planes. The inflammatory response only boils to the surface as dermal changes in sporadic areas, like peaks to a mountain range. What we see are skip lesions and tenderness far removed from these lesions.

    With necrosis of vessels in the subcutaneouslayer comes extravasation of blood into neighboring necrotic dermis. The fluid percolates to the surface, contained by the only layer of epidermis immune to a necrotizing process, a layer already dead. The outer keratin covering termed stratum corneum caps domes of bloody fluid we call hemorrhagic bullae.

    Even if we believe anchored tissue indicates containment of the organism, it may not be so. It had already gone systemic in hematogenous transit, filling the circulation with a toxin that invokes a systemic inflammatory response. These organisms are faster and more clever than we appreciate.

    Comments about this article? Write to EMN atemn@lww.com.

    Click and Connect!Access the links in this article by reading it onwww.EM-News.com.

    Image
    Image:
    Image

    Dr. Morchiis the director of the Medical Screening Examination program at Harbor UCLA Medical Center and an assistant professor of emergency medicine at UCLA's David Geffen School of Medicine.

    Only Online

    Read all of Dr. Morchi's past columns in the EM-News.com archive.

    © 2011 Lippincott Williams & Wilkins, Inc.