Journal Logo

First Person

First Person

Six Tries to Get the Diagnosis Right

Miller, Daniel G. MD; Negaard, Matt MD

Author Information
doi: 10.1097/01.EEM.0000552815.50596.52
    diagnosis, shingles, herpes zoster
    diagnosis, shingles, herpes zoster:
    Examining the patient's waistline revealed this rash in the area of her pain.

    When “Dee” checked into our ED on a busy Monday, the eye-rolling extended from triage on back to the physicians. This was her fourth visit in two weeks for the same complaint. A pleasant woman in her 60s, Dee had multiple chronic conditions, including asthma, hyperlipidemia, type 2 diabetes, and hypertension. She was also less than a year out from chemotherapy, radiation, and a hysterectomy for endometrial cancer that was complicated by a left-sided pelvic abscess. Dee had recently returned from a camping trip, and since then, she had been experiencing a nagging, sharp pain on her right side.

    She saw her primary care physician for the pain as well as chills and nausea. Her physical exam was documented as grossly normal other than right costovertebral angle tenderness and right lower quadrant pain. Her urine was positive for blood, protein, and glucose and her creatinine had increased to 1.5. Her physician suspected kidney stones and arranged for an outpatient CT urogram, which showed left hydronephrosis and hydroureter with dilation terminating at the level of the previous abscess. No associated left-sided renal stones were seen, but the right kidney had a 2 mm stone without obstruction. She received tramadol for pain and was referred to urology.

    But Dee developed nausea and vomiting after starting tramadol, so she stopped taking it the next day. The pain worsened, and she came to the ED. She was not put into a room for several hours, and was one of the night team's final patients. She reported worsening right flank pain and hematuria. Her labs and vital signs were unchanged, and an ultrasound redemonstrated the CT findings.

    Urology was consulted and found no indication for emergent surgery. They recommended straining urine, arranged for outpatient cystoscopy with left ureteral stenting, and asked that alternative causes for the patient's pain be considered. After ketorolac, ondansetron, and hydrocodone, Dee reported improved symptoms and was discharged, but she continued to have right-sided flank pain. Four days later, a left ureteral stent was placed without incident. Her pain seemed controlled on hydrocodone, but she came back to the ED for the fourth time in two weeks after her prescription ran out. The physician assistant in triage ordered a CBC, chemistry, and urinalysis, gave Dee a dose of hydrocodone, and sent her back to the waiting room for six hours.

    Then the charge nurse reported that Dee's labs were normal and her pain was much better, and asked if she can be placed in a hallway bed for discharge. Everything looks stable in recent notes and labs. What would you do?

    Look, Listen, and Feel

    We placed Dee in an available room. She described a burning sensation wrapping from the right side of her back to her groin. When we asked Dee to show us the areas where it hurt, she pulled her pants down several inches, and the diagnosis became evident: herpes zoster!

    Dee had been unaware of the rash, so she was unable to report how long the lesions had been present or whether new lesions were forming. Given her age, pain, and history, we began valacyclovir therapy. Dee had no subsequent visits, CTs, or surgical procedures for her shingles.

    This case demonstrates the importance of looking at the area that patients say hurts. Medical students are taught to look, listen, and feel, but EPs too often fail to look, especially when seeing patients in hallways. The H&P must be the basis for tests and treatment, but EM is increasingly protocolized and fragmented, diffusing responsibility for each patient's assessment. Here, a primary care physician, tired nocturnist, urologist, nurse in triage, and charge nurse were involved in the care. Satisfaction and length-of-stay metrics have been imposed in the name of quality, but their accuracy and benefit are dubious. (West J Emerg Med 2014;15[2]:170;; Arch Intern Med 2012;172[5]):405;

    Having a medical professional in triage may decrease length-of-stay and left-without-being-seen rates, but their impact on patient-centered outcomes and resident education is concerning. (Am J Emerg Med 2018;36[1]:124; Ann Emerg Med 2017;70[4]:S44; West J Emerg Med 2014;15[7]:902; Increased imaging in EDs has not corresponded with improved outcomes (Ann Emerg Med 2013;62[5]:486; PLoS One 2013;8[6]:e65669;, and hallway beds are associated with increased morbidity and mortality. (Adv Emerg Med 2014;ID 495219;

    Emergency physicians often feel buffeted by the conflicting interests of metrics, resource utilization, consultant happiness, and malpractice concerns. No wonder we sometimes forget that our primary imperative is to help the patient. We spend years building a foundation of principles upon which to base our algorithms, clinical guidelines, and checklists. This allows us to notice when a patient doesn't fit into a predetermined clinical pathway or when a diagnosis doesn't explain the symptoms. Failure to honor these foundations undermines physicians' value. We become cogs in a system loose from its foundation. If we fail to look, listen, and feel to determine if the diagnosis actually explains the patient's symptoms, we will stop being physicians and become replaceable technicians.

    Share this article on Twitter and Facebook.

    Access the links in EMN by reading this on our website or in our free iPad app, both available at

    Comments? Write to us at [email protected].

    Wolters Kluwer Health, Inc. All rights reserved.