Diagnostic Accuracy of Point-of-Care Ultrasound for Retinal Detachment

Gottlieb M, Holladay D, Peksa GD. Point-of-Care Ocular Ultrasound for the Diagnosis of Retinal Detachment: A Systematic Review and Meta-Analysis. Acad Emerg Med. 2019 Jan 13. doi: 10.1111/acem.13682

Study Population: 11 studies comprising 844 patients suspected of having retinal detachment, with 5 studies conducted in the emergency department

Ocular complaints are common in acute care settings. Although most patients will ultimately be diagnosed with a benign condition, 10-26% of people presenting to medical providers with flashes and floaters, for instance, will be diagnosed with retinal detachment.1^, 2 Rapid diagnosis and treatment of retinal detachment are crucial to prevent irreversible vision loss.3 Unfortunately, the patient’s history has been shown to be poorly predictive, and physical examination performed by a non-ophthalmologic acute-care provider has demonstrated poor positive and negative predictive value.4 The gold standard for diagnosis of retinal detachment is largely indirect fundoscopy by an ophthalmologist. Point of care ultrasound (POCUS) may provide a fast and accurate assist in diagnosis, particularly when specialist consultation for fundoscopy may be delayed or inaccessible.

The meta-analysis discussed here evaluated the test performance of POCUS for the evaluation of suspected retinal detachment.5 Prospective (randomized-controlled or not) studies of POCUS in patients suspected of retinal detachment were included with a confirmatory test (ophthalmologic examination, surgical findings, computed tomography, magnetic resonance imaging, or clinical follow up). The primary outcome was diagnostic accuracy of POCUS for retinal detachment.

The authors of the meta-analysis identified 11 observational studies (n = 844 patients) that met their inclusion criteria.5 Consistent with other population estimates,1^, 2 21% of patients were ultimately diagnosed with retinal detachment. Overall, POCUS was 94% sensitive (95% confidence interval [CI] 78% to 99%) and 96% specific (95% CI 89% to 99%), with a positive likelihood ratio (LR+) of 25 (95% CI 8 to 78) and a negative likelihood ratio (LR-) of 0.06 (95% CI 0.01 to 0.25). The area under the receiver operating characteristic curve demonstrated high accuracy (0.99; 95% CI 0.97 to 0.99).

The performance of POCUS is dependent on provider experience, and the included studies involved a range of clinician experience and training, possibly accounting for some heterogeneity. Of note, most studies had consistently high sensitivity and specificity regardless of operator specialty, with the exception of one study which included a medical student sonographer and demonstrated reduced sensitivity compared to the other included publications. Training protocols varied between studies, ranging from 30 minutes to 2 hours in length, so it is unclear what the ideal training protocol should entail. The authors of the meta-analysis were unable to evaluate accuracy based on clinician experience, as most studies included sonographers of varying experience levels.

Of 11 studies included, only 5 were conducted in Emergency Department (ED) patients. The accuracy of POCUS was high in these patients [93.9% (95% CI 78.7% to 98.5%) sensitive and 92.4% (95% CI 85.6% to 96.1%) specific], but was significantly lower in other populations [74.1% (95% CI 61.0% to 84.7%) sensitive and 85.3% (95% CI 75.3% to 92.4%) specific]. This may be due to differences in the patient populations (e.g., acuity) or location sonography. Importantly, test characteristics were comparable when truly point of care ultrasound was performed by emergency medicine physicians (6 studies) as when performed by radiologists (4 studies). Further studies are recommended to identify which subgroup characteristics are associated with decreased diagnostic accuracy in these various cohorts.

Studies were at low risk of bias overall, though reference standards were an issue. The multiple different reference standards used across studies would normally be a concerning source of potential bias and variation for pooled accuracy data, but the consistently high sensitivities and specificities across studies with different reference standards are reassuring. The confidence intervals for both pooled sensitivity and specificity were wide. This indicates that high quality studies with larger sample sizes are needed. Future research should assess the influence of sonographer experience, as well as the ideal training protocols and number of examinations to obtain proficiency in this modality. It would also be valuable to determine the effect of high-frequency linear transducers, color Doppler, and contrast-enhanced ultrasound on diagnostic accuracy, as color or focused Doppler ultrasound may be more accurate than B-mode.6

Based on the existing evidence, POCUS appears to be a rapid, accessible, noninvasive tool that is highly sensitive and specific for diagnosis of retinal detachment. In fact, the American College of Emergency Physicians lists ocular examination with POCUS as a core emergency ultrasound application.7 Therefore, we have assigned a color recommendation of Green (Benefit > Harm) for use of POCUS for diagnosis of retinal detachment. POCUS should be considered a powerful tool to expedite definitive care for patients when direct ophthalmologic consultation may be otherwise delayed or limited.

April 15, 2019

LR, pretest probability and posttest (or posterior) probability are daunting terms that describe simple concepts that we all intuitively understand.

Let's start with pretest probability: that's just a fancy term for my initial impression, before we perform whatever test it is that we're going to use.

For example, a patient with prior stents comes in sweating and clutching his chest in agony, I have a pretty high suspicion that he's having an MI – let's say, 60%. That is my pretest probability.

He immediately gets an ECG (known here as the "test") showing an obvious STEMI.

Now, I know there are some STEMI mimics, so I'm not quite 100%, but based on my experience I'm 99.5% sure that he's having an MI right now. This is my posttest probability - the new impression I have that the patient has the disease after we did our test.

And likelihood ration? That's just the name for the statistical tool that converted the pretest probability to the posttest probability - it's just a mathematical description of the strength of that test.

Using an online calculator, that means the LR+ that got me from 60% to 99.5% is 145, which is about as high an LR you can get (and the actual LR for an emergency physician who thinks an ECG shows an obvious STEMI).

(Thank you to Seth Trueger, MD for this explanation!)