emergency medicine iceland

Nov 16, 2014

Male with abdominal pain and a flabbergast passenger

This case of a gentleman with lower abdominal pain is interesting not only clinically, but more so through the prism of basic cognitive science and how it applies to clinical judgement. We take a closer look at how we are prone to making mistakes due to heuristics and cognitive biases. Also included are some gems on the obvious usefulness of ultrasound at the bedside and a gentle reminder about interpreting an elevated D-dimer result. Sit back, grab a 'kókómjólk' and enjoy the case.

Abdominal pain

51 y/o B. F. with prior history of appendectomy in childhood presented to the ED with abdominal pain after referral from his GP.

The pain was localized to the left iliac fossa and had been intermittently increasing during the last week. It was coupled with increasing abdominal distension. No acute changes noted on the day of his visit. No fever or chills. Denies nausea and was not sick in the last week. He experienced runny stool earlier in the week and was constipated for 2 days but had a normal bowel movement yesterday without any blood or mucus. No bowel changes in the last months and no urinary tract symptoms. Noticed reduced appetite, did not eat since yesterday. B. F. also believed he had been losing weight over the last couple of weeks.

On physical examination the patient looked tired but in no acute distress. Vitals were all normal. Moderately distended abdomen, diffuse tympany. Bowel sounds present but faint and distant. Diffuse pain on pressure but no tenderness or guarding.

Cue in bedside ultrasound

Bedside ultrasound of the abdomen was performed. Visualizing was complicated by substantial amounts of air in the GI tract. Liver appeared normal without any suspicious growths on gross examination. Noted significant dilation of small bowel mostly in left flank with fluid contents and hyperperistalsis (like a "nest of worms").

At this point there was a strong consideration of ileus or at least subileus and since patient had distended abdomen a feeding tube was placed but which drained only 230 ml of GI fluids. Patient received 5 mg of Morphine.

What are the diagnostic criteria for small bowel obstruction on ultrasound examination?
  • Dilated bowel loops > 25 mm
  • Increased intraluminal fluid
  • Characteristic alternating peristalsis
  • Keyboard sign (visible plicae circularis)
  • Tanga sign (circumscribed free fluid)

The sensitivity and specificity of detecting a dilated bowel with bedside US after a merely 10-minute training session is 91 and 84 %, respectively. Compared to the 50-80 % diagnostic accuracy of a plain CXR, otherwise most commonly used at the ED for establishing the diagnosis. For further info see Jang et al. - Bedside ultrasonography for detection of small bowel obstruction in the emergency department.

A short but excellent introductory video tutorial on performing the examination can be found here and is presented by the legendary Chris Fox, emergency physician with a fellowship in ultrasound and book author - he is by far one of the best ultrasound teachers and all his lectures can be found online for free - FOAM at it's best! The small bowel exam starts at 13:30.

An informative and more academic post on SBO and the whole ultrasound vs other modalities, even a CT/MRI scan, can be found on Academic Life in Emergency Medicine, an incredibly useful site in general.

Putting the pieces together

Blood panel came back normal (WBC 5.000) aside from an elevation of CRP at 60 mg/L. Liver enzymes normal and urine dipstick test clean.

Because of suspected ileus/subileus patient was sent for an abdominal CT scan to look for underlying cause (remember, ultrasound is excellent for finding ileus but bad for finding the cause).

Radiologist's answer for the CT image was that there was no ileus, no free gas and no free fluid in abdomen. The clinical scenario was not on par with the radiologist's answer so the images were reviewed and discussed with the on-call surgeon. We agreed there were indeed partially dilated small bowel loops, "sentinel loops" and the surgeon was even concerned about some mesenterial veins being prominent, possibly suggest mesenteric venous thrombosis (MVT).

What are sentinel loops?
A sentinel loop is a focal area of adynamic ileus close to an intra-abdominal inflammatory process. The sentinel loop sign may aid in localizing the source of inflammation. For example, a sentinel loop in the upper abdomen may indicate pancreatitis, while one in the right lower quadrant may be due to appendicitis.
For more go to Radiopedia or RadiologyMasterClass.

CT abdomen; frontal, coronal and sagittal views

Incidentally the prostate is calcified.

Are prostatic calcifications always pathologic?
No. Prostatic calcification is most often an incidental, asymptomatic finding but it can cause symptoms such as dysuria, infection, haemautira, obstruction or pelvic/perineal pain. Occasionally calcifications can be passed via the urethra.
    Prostatic calcification may be either primary (idiopathic) or secondary to:
  • diabetes mellitus
  • infections - e.g. tuberculosis or bacterial prostatitis
  • benign prostatic hypertrophy - calcification occurs in 10%
  • prostate cancer
  • radiation therapy
  • iatrogenic - urethral stents or surgery
Radiopedia has more information on prostatic calcification.

A d-dimer was ordered which came back 2.890 ng/mL. A venous blood gas was also taken but came back within expected limits.

What are the possible pathological and non-pathological causes of D-dimer elevation
    Pathological possibilities are manyfold and can make life harder for us when we order it beyond its intended use:
  • Acute coronary syndromes
  • Acute upper gastrointestinal haemorrhage
  • Aortic dissection
  • Arterial or venous thromboembolism
  • Atrial fibrillation
  • Consumptive coagulopathy – DIC, VICC
  • Infection
  • Malignancy
  • Pre-eclampsia
  • Sickle cell disease
  • Stroke
  • Superficial thrombophlebitis
  • Trauma
  • On the not-too-pathological spectrum are the following standard conditions:
  • Age
  • Cigarette smoking
  • Functional impairment
  • Post-operatively
  • Pregnancy
  • Race

A D-dimer should be ordered when we have a low pre-test probability for venous thromboembolism (VTE). The latter can be assessed with any of the scores available, usually the Wells criteria for DVT, which can be found on MDCalc.

Gotcha ...

As everybody's eyes were focused on the dilated bowels an important find was missed until the images were reviewed for the second time… a large fluid-filled urinary bladder measuring about 15x12x7 cm - even the radiologist made no mention of it!

B. F. was diagnosed with urinary retention and catheterised in the ED with 1.500ml of clear urine passing in a few hours.

It is interesting to note that B. F. passed urine about 2 hours before the CT study was performed and still had about 1000ml of fluid visible on the CT scan. His HPI also did not include any urinary tract symptoms but upon further inquiry he describes hematospermia but had not seen a doctor for this.

B. F. was admitted to general surgery for observation but as symptoms relieved the suspicion of MVT was aborted. Urology was consulted and patient diagnosed with overflow incontinence. Patient was sent home with a urinary catheter for 10 days and Tamsulosin 0.4 mg x 1.

At outpatient follow-up no cause was found for his retention - the prostate was described as normal on palpation and ultrasound and cystoscopy showed no pathology. Patient continued on Tamsulosin and had no further urinary events but noticed memory problems and was referred for further cognitive evaluation.

What clinical entity could explain this?
Non-pressure hydrocephalus - although unlikely since gait and memory problems normally present before incontinence.
What causes overflow incontinence?
Transient urinary incontinence is often seen in both elderly and hospitalized patients. The mnemonic DIAPPERS is a good way to remember most of the reversible causes of incontinence, as follows:
  • Delirium or acute confusion
  • Infection (symptomatic UTI)
  • Atrophic vaginitis or urethritis
  • Pharmaceutical agents
  • Psychological disorders (depression, behavioral disturbances)
  • Excess urine output (due to excess fluid intake, alcohol, caffeine, diuretics, peripheral edema, CHF, hyperglycemia or hypercalcemia)
  • Restricted mobility (limits ability to reach a bathroom in time)
  • Stool impaction

To pee or not to pee

Urinary incontinence is defined as the involuntary loss of urine that represents a hygienic or social problem to the individual. Urinary incontinence can be thought of as a symptom as reported by the patient, as a sign that is demonstrable on examination, and as a disorder. Urinary incontinence should not be thought of as a disease, because no specific etiology exists. The etiologies of urinary incontinence outlined above are diverse and incompletely understood.

Patients with urinary incontinence should undergo a basic evaluation including a history, physical exam and a quick urinalysis.

Here you can take a look at a nice drawing comparing different types of incontinence.

It’s important to note the dynamics of urine loss with overflow incontinence since the bladder isn’t just filled to the brim and then it just empties at a certain point.

On the contrary, the bladder constantly remains full but just leaks small amounts of urine over time. Patients might not even notice any abnormalities, as was the case with B. F. The basic workup is aimed at identifying possible reversible causes. If no reversible cause is identified, then the incontinence is considered chronic. The next step is to determine the type of incontinence (urge, stress, overflow, mixed, or functional) and the urgency with which it should be treated. With slowly developing or chronic obstructions, patients typically are older, with multiple comorbid conditions, and they present with overflow incontinence and report little to no pain. But all that being taken into account, this is usually not a problem for the ED and should be handled by the patient’s GP.

To err is human a.k.a. discussion

Now, the reader that made it all the way through might come to ask the question: “Why’d they chase the ileus diagnosis after doing an ultrasound, when they could’ve seen the bladder?”

This is a very valid question that opens up a further discussion of cognitive errors in clinical judgement. While doing bedside US the examiner might have gotten caught up in a possible diagnosis of ileus, focusing most of his attention on this eventuality. Once they noted a possible distended loop of bowel, they unknowingly paid less attention to the rest of the systems that might have been involved. This in turn led everyone to temporarily miss the overexpanded bladder that would have been clearly visible on US as well. This is called anchoring.

Anchoring occurs when we sink our teeth into a possible diagnosis and pursue it, while discarding other possibilities. It is a type of thinking much akin to the “horses and zebras” analogy.

It is the tendency to fixate on specific features of a presentation too early in the diagnostic process and to base the likelihood of a particular event on information available at the outset. This may often be an effective strategy. However, this initial impression exerts an overly powerful effect in some people and they fail to adjust it sufficiently in the light of later information. Anchoring can be particularly devastating when combined with confirmation bias.

Anchoring may lead to a premature closure of thinking. Patients may be labeled with an incorrect diagnosis very early on in their presentation. Diagnoses, once attached, are difficult to remove and may seal the patient’s fate. (Croskerry, 2002)

Heuristics? Metacognition? Reflection ... Argh, where’s my Ritalin? Just tell me how I can fix it!

Luckily some smart people have identified strategies to battle heuristics and cognitive biases.

First and foremost, as with all types of bias, we need to be aware at all times that we are prone to biased decision-making. This is called metacognition and it has been proven effective in the past.

Checklists. Checklists reduce cognitive load and can improve patient outcomes and safety. A properly structured and intelligently designed patient interview form that is thoroughly implemented in the work environment can help us avoid missing an important step. A very cool and accessible book on the subject by dr. Atul Gawande is The Checklist Manifesto.

Morbidity and mortality conferences are helpful in highlighting cases where cognitive error was the source of raised morbidity.

Clinical decision support (CDS) systems are emerging as a novel adjuvant to the classic approach to the clinical decision process. Modern solutions are integrated into the electronic health record (EHR) and can provide weighted suggestions to clinicians, based on multiparametric inputs. (Dinevski, 2011) One of the core objectives of the Meaningful Use project in the United States, that aims to incentivize EHR adoption is implementation of at least one CDS system.

Recent additions to the body of knowledge on this subject however emphasize the importance of a wider change of culture that must take place in order to not only reduce errors but to also become aware of them in a healthy manner that does not promote defensive medicine.

Four take home messages

1. D-dimer can tell you some things, but you wanna know which. In this case a follow-up measurement might be warranted to establish what the cause of the elevated D-dimer was, since ileus secondary to thrombosis was ruled out. Or was it ... (a CT venography was never done!)

2. Ultrasound is cool for diagnosing ileus. High specificity and sensitivity compared to plain film. Can be performed at the bedside in the emergency setting. Fast, cheap, no radiation. Need we say more?

3. We tend to make cognitive errors. Let’s not forget it. And with a bit of practice and vigilance we can avoid the most common pitfalls. We outlined some of the strategies above.

4. Overflow incontinence presents as otherwise asymptomatic lower abdominal pain. This is attributed to the fact that there is no de facto retention since patients pass urine “normally”. The patient will present with discomfort in the abdomen. The thorough and systematic ultrasound examination is key to establishing the proper diagnosis in such cases. Plus it seems wise to keep urinary retention somewhere on the roster of differentials for lower abdominal pain with both genders.

4. Overflow incontinence presents as otherwise asymptomatic lower abdominal pain. This is attributed to the fact that there is no de facto retention since patients pass urine “normally”. The patient will present with discomfort in the abdomen. The thorough and systematic ultrasound examination is key to establishing the proper diagnosis in such cases. Plus it seems wise to keep urinary retention somewhere on the roster of differentials for lower abdominal pain with both genders.

Further suggested reading

D-dimer

Ultrasound for SBO

Overflow incontinence

Clinical cognitive biases

Author: Jan Hansel, a 6th year med student from Slovenia, passionate about medical education, emergency medicine and music. A young FOAMer in training. [Editor's note: Jan visited our ED in August 2014 and not only practiced emergency medicine like a boss but also tried swimming in the cold Atlantic!]