Diagnostic Accuracy of Non-Invasive Thermal Evaluation of Ventriculoperitoneal Shunt Flow in Shunt Malfunction: A Prospective, Multi-Site, Operator-Blinded Study
- Jul 09, 2020
Galal Elsayed, MD and Brandon Rocque, MD, Birmingham,AL
- Prospective, multicenter, non-randomized, operator-blinded trial of consecutive series of suspected shunt malfunction patients with non-invasive thermal flow evaluation (TFE) of a ventriculoperitoneal shunt versus neuro-imaging in clinic or the emergency department.
- Inclusion criteria: Patients between ages of 3-29, repeat patients were enrolled again if presentations were separated by more than 1 month or if surgical intervention occurred. Exclusion criteria: patients with multiple catheters, edema over the shunt, catheters were not palpable over the clavicle or where testing interfered with clinical care.
- 406 patients were enrolled at 10 pediatric tertiary centers in North America, TFE performed on 391 patients
- Reference standard for shunt failure was shunt revision surgery with visually confirmed shunt obstruction within 7 days of TFE.
- A total of 77 patients progressed to surgery for visually confirmed shunt obstruction
- Cohorts: TFE with automatic micro shunt valve pumper (MP) and TFE alone: nearly all of them received neuro-imaging.
- Results: TFE-MP among all patients: Sensitivity 90%, Specificity 50%, NPV 95%, PPV 30%, LR+ 1.8, LR- 0.2. Neuroimaging vs TFE-MP among all patients: Sensitivity 76%, Specificity 90%, NPV 94%, PPV 66%, LR+ 7.8, LR- 0.3. TFE-only among all patients: Sensitivity 100%, Specificity 60%, NPV 100%, PPV 29%, LR+ 2.5, LR- 0. Neuroimaging vs TFE-only among all patients: Sensitivity 77%, Specificity 92%, NPV 96%, PPV 61%, LR+ 9.5, LR- 0.3.
- Thermally detectable flow predicts non-progression to surgery for obstruction and is non-inferior to neuroimaging since 95% confidence intervals for TFE did not cross non-inferiority margin versus neuroimaging.
- Neuro-imaging has higher specificity compared TFE but diagnostic accuracy of TFE is higher among those with preimaging clinical impression of “unlikely to require shunt revision”
- Given these results, TFE may be an excellent “rule-out” test for those clinically classified as “low risk” and may be considered for discharge without neuro-imaging.