The Accuracy of POCUS for Diverticulitis

By Michael Prats

Diverticulitis 101

A Prospective Evaluation of Point-of-Care Ultrasonographic Diagnosis of Diverticulitis in the Emergency Department

Ann Emerg Med 2020 Jul (Epub) - Pubmed Link


Take Home Points

1. In the largest prospective trial to date, POCUS in the hands of highly trained operators had high specificity and fairly high sensitivity for diverticulitis compared to CT.

2. More studies are needed to show the potential patient-centered benefits of a diagnostic algorithm that incorporates POCUS.


Background

Diverticulitis is a common cause of acute abdominal complaints. Sometimes the presentation can be vague or atypical, making it challenging to diagnose. Unfortunately, it can progress to serious complications, so it is important to make the diagnosis and initiate treatment in a timely manner. If the patient needs imaging (certainly, not everyone does), most will reach for their trusty CT scanner. However, we know that this leads to radiation and can significantly increase the length of stay in the emergency department. So what about point-of-care ultrasound? This has been described before, although not well studied in this population. If ultrasound was accurate for this diagnosis, it might be a great choice for many patients who could then be treated and sent home. This study compares a POCUS scan for diverticulitis to the reference standard CT seeking to evaluate the accuracy of this strategy.


Question

What are the test characteristics of point-of-care ultrasonography in the diagnosis of diverticulitis in the ED?


Population

Patients in the emergency department, data collected over 2.33 years


Inclusion:

  • Abdominal pain with suspicion by treating physician for diverticulitis and plan to order CT

  • Trained provider available to enroll


Exclusion:

  • Clinical instability

  • Pregnancy

  • ≤17 years old

  • Abdominal surgery within 2 weeks

  • Pre-confirmed diagnosis of diverticulitis

  • Unable to consent

  • No CT scan performed


Design

Prospective, convenience sample, single center

The sonographers were blinded to patient clinical data and imaging

Treating physicians were blinded to the results of the POCUS

Study personnel collected information from medical records such as demographics, previous abdominal surgeries, prior diverticulitis, height, weight, and disposition


Ultrasound was performed prior to CT (see below for protocol)

All patients in the study received CT (it was an inclusion criteria) CTs could be with or without contrast depending on treating physician discretion

The POCUS was compared to the criterion standard of CT. Interpretation of CT was by attending radiologist. Considered positive if “diverticulitis” was listed in impression of report.


Primary outcome measures were sensitivity, specificity, positive predictive value, negative predictive value.

Power calculation - estimated they would need 406 patients (assumed 18% incidence, 95% sens, 95% spec)


Who did the ultrasounds?

Ultrasound fellows or fellowship-trained emergency physicians or physician assistants

40 minute didactic training course reviewing criteria and study protocol, 5 precepted scans (at least 1 positive) prior to enrolling


The Scan

Curvilinear transducer

Curvilinear


Initiated at point of maximum tenderness

Graded compression to visualize bowel

“Lawn mower” technique to scan entire abdomen


Criteria for Positive POCUS - needed ALL of the following in the same location:

  • Bowel wall edema > 5 mm surrounding a diverticula

  • Enhancement of pericolonic fat

  • Sonographic tenderness to palpation


Learn how to do POCUS for Diverticulitis from friend of the show Joe Minardi

Here's another good example from Dr. Gerald Diaz on GrepMed


Results

N = 452

  • 4 were excluded for “unable to complete POCUS” (unclear why)

  • Median age 60

  • 54% women

  • Median BMI 27.4

  • 67% had left lower quadrant pain

  • 33.85% had diverticulitis in the past

  • 36% of patients had diverticulitis on CT

  • 36% of the diverticulitis patients were admitted


Primary Outcome - Test characteristics of POCUS compared to CT for diagnosis of diverticulitis

Sensitivity 0.92 (0.88–0.96)

Specificity 0.97 (0.94–0.99)

LR+ 30.67

LR- 0.08


Additional Findings

Mean time to complete POCUS 4.9 minutes (SD 1.9 minutes)

All of the patients with complicated diverticulitis (listed as 35/161 in two places, but 37/163 in discussion section) had sonographic diverticulitis. Also sometimes had free fluid or abscess seen on ultrasound.


Strengths

Prospective study of emergency department patients

Fairly large population, met power analysis

Appropriate statistics

High degree of accuracy, short time of examination


Limitations

Selection bias from convenience population. Authors state that the majority of patients were enrolled during weekday business hours.

Highly trained sonographers. Majority of scans were performed by 5 ultrasound faculty members.

The choice in criteria for a positive scan may have led to decreased sensitivity and increased specificity. A cited article states that diverticula are only seen on ultrasound in 50% of cases. It makes sense to include pain over area of the sonographic findings, but there are scenarios when that might not be the case. The bowel wall thickness cut-off was made 5 mm, although >3mm is technically abnormal.

The population included only those cases in which a CT scan was performed. This could be a distinct subset of the population of patients with acute diverticulitis. Perhaps it was because they were sicker or higher risk for complication that led to the decision to obtain CT. These results might not apply to all comers. This is supported by the fact that a fairly large proportion of the patients had diverticulitis and 36% were admitted with diverticulitis.


Discussion

Another fairly large, similar article recently published (July 2020, Ultraschall Med)

  • Multicenter, POCUS for diverticulitis in 4 EDs, all patients had CT in ED and at 1 month followed up. Reference standard was chart review including CT imaging.

  • More varied sonographers - IM, surgery faculty, not fellowship trained, did have 2 hour training. They used curvilinear but followed it up with a linear look at bowel wall.

  • Their criteria were >4mm thickening, otherwise similar. Great videos of pathology.

  • Results: N = 393, 55% had diverticulitis. POCUS saved time to diagnosis (by about 2 hours), 92.7% sensitivity, 90.9% specific.

  • Overall similar study but does provide some additional information. Broken down by each individual component of the scan - thickened bowel wall was most sensitive ~80%. Pericolonic air foci were 100% specific, after that the most specific was echogenic fat at 92.6% specific. Only 50% sensitive for detecting complicated diverticulitis.

Interestingly (back to the main study) the false positives were mostly (7/10) due to CT-confirmed colitis with coexisting diverticulosis. There were 13 false negatives, all had uncomplicated diverticulitis.

Let's think about down the road how POCUS could be integrated into the clinical workup for suspected diverticulitis. It really depends on your current practice. If you are getting CT’s only when you suspect complicated diverticulitis, then it might not change too much because you probably still need a CT for that. If you are getting CT’s for all of your diverticulitis, you could potentially stop that and use ultrasound (if you are skilled and have had some practice). Certainly, given that this exam seems fairly specific - if you have a reasonably moderate to high pre-test probability + POCUS findings of diverticulitis + low suspicion for complicated diverticulitis → might be reasonable to treat without confirmatory imaging. What do you think?


Summary

This is a prospective study in the emergency department enrolling patients with suspected diverticulitis, getting a CT. They performed a POCUS protocol that took about 5 minutes and yielded a 92% sensitivity, and 97% specificity compared to CT.


Take Home Points

1. In the largest prospective trial to date, POCUS in the hands of highly trained operators had high specificity and fairly high sensitivity for diverticulitis compared to CT.

2. More studies are needed to show the potential patient-centered benefits of a diagnostic algorithm that incorporates POCUS.


Our score

4 Probes


Expert Reviewer for this Post

Au

Arthur Au, MD @arthurkau

Arthur is the Associate Ultrasound Director and Ultrasound Fellowship Director for the Department of Emergency Medicine at Thomas Jefferson University in Philadelphia, PA.


Reviewer's Comments

Overall this was a well-designed study that suggests that POCUS may be a reasonable alternative to CT in the initial evaluation of patients with concern for diverticulitis. While the providers in this study were certainly highly trained in POCUS, the limited additional training and short, straightforward scanning protocol makes me believe that this exam could be learned by most providers who are already facile in POCUS. Incorporating this into practice could be similar to the use of POCUS in renal colic. If the presentation and ultrasound findings are both suggestive of an uncomplicated diverticulitis, it’s likely reasonable to give these patients a trial of medical management and forgo CT imaging. Given the potential time savings and decrease in radiation exposure to patients, I’m excited to see how this plays out in clinical practice.


Cite this post as

Michael Prats. The Accuracy of POCUS for Diverticulitis. Ultrasound G.E.L. Podcast Blog. Published on November 09, 2020. Accessed on December 05, 2020. Available at https://www.ultrasoundgel.org/101.
Published on 11/09/20 04:00 AM
comments (0)