Journal Logo

Diagnosis Deconstructed

Diagnosis Deconstructed: Haunted by Incidentalomas

Morchi, Ravi MD

Author Information
doi: 10.1097/01.EEM.0000431340.30398.4f
    Figure
    Figure
    Figure
    Figure

    At signout rounds, I hear about a 55-year-old with Down syndrome brought in by paramedics for near syncope an hour earlier. His eyes rolled up “into the back of his head,” and he began “breathing funny” at home. It is unclear if he truly lost consciousness. He had no motor activity. The event ended as abruptly as it began, and his only complaint afterwards was diffuse abdominal pain.

    His mentation was at baseline upon arrival, and the initial exam was significant for diffuse abdominal tenderness with fecal and urinary incontinence.

    The paramedic 12-lead and the initial ED ECG showed low voltages and T-wave flattening in the precordial leads. An abdominal CT scan had been performed to “make sure there's nothing going on in there.” I know there is plenty going on in there … peristalsis, biliary contractions, exocytosis, endocytosis, brush border enzyme activity, translocation of proteins and fats across plasma membranes, gastrin release, serotonin production, negative and positive feedback loops, cellular respiration, an enteric intramural basal electrical rhythm. And that is just the beginning.

    Just whatexactly are we looking for? CTs done without a clear differential diagnosis in mind can complicate matters, generating incidental findings that float in cyberspace because they lack a clinical anchor. Scans ordered based on a formula: “elderly + abd pain = CT” or “immunocompromised + abd pain = CT” or “third visit for the same + abd pain = CT.” But we still need a differential diagnosis. Simply reaching a threshold to order imaging is not enough; the burden rests with us to create and weigh the possibilities in our minds.

    Based on the locale of tenderness and the depth of involuntary, non-distractible guarding. Reflexive contraction of the rectus and oblique musculature meant to prevent our hand from dipping the parietal peritoneum against an inflamed structure deep to it. Contraction that is consistent in location, depth, and timing with every push.

    Based on the absence of such peritoneal irritation. Maybe the lining cannot abut an inflamed structure because of a thick greater omentum coating and sequestering the intraperitoneal process. Or because the structure in question is in the retroperitoneum, or below its inferior reflection deep within the true pelvis.

    Based on the quality of pain. Sudden onset with an immediate rise to maximum intensity, only to fall off and recur at intermittent intervals? That is colic. The obstruction of a tube … small bowel, large bowel, ureter, neck of the gallbladder, or CBD.

    Sudden onset with a sharp rise to maximum and an intensity that never lets up? This could be obstruction still, but add another set of tubes to the list: vasculature. And consider perforation of such tubes releasing bowel contents, biliary juice, urine, or blood.

    Gradual onset that marches forward slowly over hours and days? That is inflammatory. Obstruction to the base of the appendix or a diverticulum may have started the process, but its tempo is characterized most by the migration of neutrophils and production of local inflammatory mediators.

    The location of that tube? The address of that inflammatory activity? That depends on our hands. Where are we probing with our fingertips? If we believe the problem is intraperitoneal, we have to remove the dermis, subcutaneous tissue, and abdominal muscles in our minds, move beyond “X quadrant,” and ask, “Just what organ are we touching now?”

    If truly diffuse, do we believe that debris has been liberated into the peritoneal cavity? Exudate marinating under the parietal peritoneum, inflaming it to the extent that any motion of the lining whatsoever results in contraction of wall muscles? Does the patient really have peritonitis?

    I am scrolling through the imaging, and there is no free fluid or free air. It is unlikely that his exam would be consistent with peritonitis in the classic sense. But immediately I see there are other problems. His CT is haunted. Haunted by incidentalomas.

    The first specter I encounter is a large renal cyst. Is this the cause of his pain? Does he have deep-seated tenderness without guarding from this structure situated behind the curtain of retroperitoneum and so destined to never irritate the anterior parietal lining? Does he have CVA tenderness?

    Second, a thickened duodenal bulb. Peristalsis photographed at the wrong moment creating the illusion of thickening within the wall? Or real duodenitis or duodenal ulcer? It depends on our hands.

    Third, a small pericardial effusion. Is tenderness most marked high in the epigastrum just below his xyphoid? Am I actually pushing up against inflamed pericardial lining? Does he have a pericardial rub? His ECG is not yet consistent with pericarditis, but low voltages may be from the dampening effect of an effusion.

    The fourth apparition is a hiatal hernia. The gastric fundus lies in his thoracic cavity. It is not a paraesophageal hiatal hernia because the esophagogastric junction is also in the chest. It appears instead to be a sliding hiatal hernia. This should not be the problem because they rarely incarcerate. But is there an associated esophagitis or an esophageal ulcer unseen by CT? Would curling our fingers into his high epigastrum prove uneventful?

    Fifth, the IVC is perfectly round, not oval. Is the pressure high within it? This should not produce much abdominal pain, although downstream venous congestion of the portal system, which has to drain into the proximal IVC via the liver, could produce a sense of discomfort in the abdomen, I suppose.

    Sixth, a thickened pleura on the right. Is his abdominal pain actually chest pain? As we palpate deep in the upper quadrants, indenting the liver or gastric fundus into the diaphragm can push up against an inflamed pleural lining. In this manner, lower lobe pulmonary pathology has our patient's mouth moving and saying the word “belly.” We parrot this back at the radiologist, and a thousand pixels later we have only a few cuts through the lower thoracic cavity, and a lot of background noise from the peritoneal cavity.

    And the seventh ghostly incidentaloma? I am not sure if it is even real. The radiologist would not mention this banshee on the official read. But the finding I see concerns me.

    There certainly is a lot of paranormal activity in this study. So what does any of it mean? It means we have to go back to the bedside and do what should have been done before digitization.

    Deconstruction. He is worried, scared, and uncooperative with the exam. He is a Down's patient and his caregiver is not present at this time. We give him a small dose of sedative to get where we need to be.

    Effortless tachypnea at 24. Lungs are clear, and there is no impediment to exhalation. Is this all from pain? Tachycardic at 100, his pulse is not bounding or aggressive. PMI is difficult to appreciate. He is small in stature so maybe 91/64 mm Hg is his baseline blood pressure. He does not feel warm. Removing the 2 liter nasal cannula we find he dips from 100% to 92% and hovers there. Hypoventilatory hypoxemia from shallow respirations secondary to pain and an added sedative effect?

    Abdominal excursion during inspiration is normal. He moves in the cot without pain. Tenderness in a tympanitic area just below the liver edge. He guards consistently but only upon deep palpation. This is the distal stomach and duodenum. Nothing superficial to suggest the serosa of transverse colon. No discomfort when prodding between axillary lines in the flank. CVA percussion is unremarkable.

    We perform a bedside ultrasound of his gallbladder. No evidence for cholecystitis and the tenderness seems inferior to the fundus. There is no free fluid in Morrison's pouch.

    He said “belly,” but there is nothing on his abdominal exam that can compete with his tachypnea, nothing to mandate this degree of ventilation.

    Should we be looking in an area hidden from our palpating hand? Respiratory compensation for early ischemia confined to bowel mucosa? Acidosis from the bowel lumen or elsewhere?

    ABG 7.42, pCO2 32, pO2 53 on room air. No significant metabolic acidosis. But two important clues. The first, an A-a gradient. Aspiration associated with the syncopal event? Atelectatic alveoli? The second, a pCO2 of 32. This may appear somewhat normal on first glance, but I submit that it is not normal if your respiratory rate is 24. In that case, I would expect a lower pCO2 provided you are actually moving fresh air past alveoli.

    Our options now are relative CO2 retention, anatomic dead space ventilation, or alveolar dead space ventilation. The first would be associated with the prolonged expiratory phase and wheeze of a tiring asthmatic. The second would require shallow breaths so that inspired air is simply backwashing within a tracheobronchial tree lacking capillary interfaces, and so is unable to participate in gas exchange. The third is most concerning, and would mean fresh air is getting to alveoli, but there is not substantial blood flow to provide CO2 for deportation into the alveolar lumen.

    Ventilation but not perfusion? Intracardiac right-to-left flow can be seen in Down's patients with atrioventricular septal defects and Eisenmenger physiology, but this would not be a new problem and may not correct easily with supplemental oxygen. We need to consider an acute, intrapulmonary cause of pathologic alveolar dead space.

    He already received IV contrast to determine if anything was going on in his abdomen, so we did not scan his chest at that time. We added a D-dimer and after it returned at 2300, ensured he had no melena from a duodenal bulb ulceration, and then empirically started heparin.

    After IV fluid hydration as an inpatient, the medical team performed the next CT with the leading diagnosis in mind, vindicating the seventh incidentaloma noted earlier.

    Inadequate contrast delineation of a small, basal pulmonary vascular tributary on the original abdominal CT was verified as real by the formal CT pulmonary angiogram. He had filling defects to the right and left main pulmonary arteries, multiple peripheral pulmonary vascular occlusions, and a secondarily thickened pleura. The saddle embolus was associated with a small reactive pericardial effusion. RV strain by subsequent echo corroborated his rotund IVC on the first scan. He underwent surgical embolectomy and IVC filter placement. He was eventually discharged with preserved cardiopulmonary function status after exorcism of four of seven incidentalomas.

    Click and Connect!Access the links in EMN by reading this issue on our website or in our iPad app, both available onwww.EM-News.com.

    FastLinks

    Copyright © 2013 Wolters Kluwer Health, Inc. All rights reserved.