Why we do what we do has returned – this time focusing on the use of ultrasound in appendicitis, which has become the test de rigeur these days in most Pediatric Emergency Departments. I won’t belabor the point on how important it is to correctly diagnose appendicitis. It peaks between ages 9-12, and can lead to perforation within 36-72 hours. Missed appendicitis is also one of the biggest causes of filed malpractice claims.

How is it typically performed?

The most common technique is called graded compression. A linear probe is placed over the point of maximal tenderness and the bowel contents are pushed out of the way. Landmarks sought include the iliac vessels and posts muscle – to which the appendix is anterior.

A normal appendix is <6mm in diameter, compressible and peristalses. The wall is generally <2 mm thick. An enlarged, non compressible appendix may also have increased flow on Doppler. Secondary signs of appendix include free fluid, a phlegmon or abscess, thickening of the mesentery and inflamed fat. If more than one of these are present the sensitivity of the ultrasound is increased. In experienced hands the appendix is seen up to 80% of the time. Pediatric Emergency Medicine and Emergency Medicine physicians can be taught to do it, with reported sensitivity in the 60-80% range after a single didactic session.

What are the possible results?

  • Normal appendix seen
  • Abnormal appendix seen
  • Appendix not visualized, no secondary signs of appendicitis
  • Appendix not visualized, one or more secondary signs of appendicitis

How good is it?

It turns out that it varies, but in general ultrasound, when performed by an experienced monographer is pretty darn good. Not CT good, but, given the lack of radiation exposure good enough to consider the first line in pediatric imaging for acute appendicitis. In general the longer the duration of pain (see Bachur et al), or the more experienced the sonographer – the more accurate the results.

The overall sensitivity 74-100%

The overall specificity 88-99%

Let’s take a look at some of the evidence in detail:

Ultrasonography and limited computed tomography in the diagnosis and management of appendicitis in children.
Garcia Peña et al. JAMA. 1999

Prospective cohort of 139 children, 108 of which had an equivocal ultrasound and then got a CT. This protocol was 94% sensitive, 94% specific and had a PPV of 90% and NPV of 97%. A negative ultrasonography result did not change the surgeons’ clinical confidence level in excluding appendicitis (P= 0.06), while, as expected a negative CT did (P<0.001).

Accuracy of noncompressive sonography of children with appendicitis according to the potential positions of the appendix
Baldisserotto et al. Am J Roentgenol. 2000

This study investigated both noncompressive and compressive techniques and saw had a sensitivity 98.5%, specificity of 98.2%, PPV 98% and NPV 98.7%.

The impact of ultrasound examinations on the management of children with suspected appendicitis: a 3-year analysis.
Dilley et al. J Pediatr Surg, 2001

The authors reviewed over 1,000 pathology reports and noted that of the 84% that actually had appendicitis 58% had received an ultrasound. Overall there was an 89% sensitivity, 95% specificity, 86% PPV, 96% NPV.

Performance of ultrasound in the diagnosis of appendicitis in children in a multicenter cohort
Mittal et al., Acad Emerg Med, 2013

A secondary analysis of a multicenter cohort from PEMCRC Of 2,625 patients with abdominal pain. 965 (36.8%) had an US – with 72.5% (95% CI = 58.8% to 86.3%) sensitivity and a 97.0% (95% CI = 96.2% to 97.9%) specificity. This takes into account both definitive and inconclusive ultrasounds.

When the appendix was clearly visualized – (48.6% of cases) – the sensitivity was 97.9% (95% CI = 95.2% to 99.9%), and the specificity was 91.7% (95% CI = 86.7% to 96.7%).

What does it mean if the appendix isn’t seen?

Recall those secondary signs mentioned above? Well, there is some evidence to suggest that they are of value even if the appendix is not seen. Estey et al in Pediatric Emergency Care in 2013 conducted a retrospective case review of 662 consecutive children with  suspected appendicitis. The appendix was not visualized in 38% (241). The additional findings considered included;

  • Free fluid
  • Phlegmon
  • Pericecal inflammatory fat changes
  • Any free fluids with prominent lymph nodes

Per the authors the odds of appy “increases from 0.56 to 0.64 to 2.3 and 17.5, respectively, when there were 2 and 3 ultrasonographic inflammatory markers identified.”

Weirsma et al in 2009 noted that in their population hyper echoic mesenteric fat was the most common secondary sign. They also noted that the NPV of not visualizing the appendix, but also have no secondary signs was close to 100%. The two studies noted above were somewhat limited by number of subjects.

The most complete study on this topic comes from Ross et al, from Academic Emergency Medicine in 2014. The authors retrospectively reviewed 968 children that had an ultrasound. The appendix was seen in less than half (45.7%, 442). 60% of those in which the appendix was not visualized were discharged home. Clinically well appearing (low suspicion patients, non-concerning exam) that were discharged home rarely had happy (0.3%). Of those discharged home based on clinical findings after incompletely visualized appendices, only 1/311 ended up having appendicitis. Ultimately 15.6% of children with incompletely visualized appendices had pathology-confirmed appendicitis – many had secondary signs.

Does anything decrease the accuracy?

It is also important to note that body habitus (BMI ≥ 85th percentile) and low pretest probability (you don’t think it was likely to be an appy to begin with) were associated with high rates of inaccurate results (interestingly more false positives!).

So, I should go with ultrasound first?

Yup, in most cases yes. Here are some specific scenarios:

  • In highly suspicious patients with a clinical diagnosis of appendicitis (teenage males) consider consulting the surgeon directly
  • If the appendix is seen and is normal, you have ruled it out.
  • Conversely, an ultrasound consistent with appendicitis means the surgeon should get a call post haste.
  • If the appendix is not visualized, but there are no secondary signs of appendicitis you have about a 1/25 chance of this patient actually ending up with appendicitis. This is probably lower in patients with a non-concerning exam – but then why are you imaging in the first place? In very suspicious cases consider a consult to surgery and/or a contrast CT scan, or even MRI – this is the so-called staged approach. Ross et al in 2014 (see above) found that the success rate of serial ultrasounds was 21% – thus, consider another imaging modality if the ultrasound was non-diagnostic.
  • If the appendix is not visualized, but the patient has one or more secondary signs of appendicitis call the surgeon. Many patients will go to the OR, others will get admitted for serial exams, and others will get a contrast CT. The important thing is not to discharge these kids home.