Shuntogram utility in predicting future shunt failures

Shunt failure remains a challenging diagnosis for neurosurgeons, particularly when patient symptoms suggest shunt malfunction but radiographic evidence is lacking. In such situations, shuntograms are sometimes employed to guide medical decision-making. In this study, the authors aimed to investigate the utility of shuntograms in aiding patient management, particularly in terms of a negative result.

This retrospective single-institution series comprises patients who underwent a shuntogram procedure to evaluate shunt patency over a roughly 6-year period. The medical records of patients were reviewed to determine the findings of the shuntogram procedure, the type of obstruction, and whether a subsequent operation for a shunt revision took place either within 30 days or up to 1 year after the procedure. Statistical analysis was completed by calculating the sensitivity, specificity, negative predictive value (NPV), and positive predictive value (PPV) of the shuntogram as compared with the revision surgery.

Of the 98 patients who underwent a shuntogram, 95 patients were included in the statistical analysis. A ventriculoperitoneal shunt was assessed in 81% of patients (77/95). The PPV of the procedure was 100%. The NPV for predicting a subsequent shunt revision within 30 days was 68.3% (false-negative rate of 31.7%) or within 365 days was 61.2% (false-negative rate of 38.8%). The sensitivity and specificity of the shuntogram were 55.8% and 100% within 30 days and 51.9% and 100% within 365 days, respectively. The most common intervention at the time of surgery following a negative shuntogram was a valve replacement in 38.5% of patients (10/26).

With an NPV of 68.3% for predicting revision within 30 days in our series of 95 patients, shuntograms remained a useful tool in the clinical decision-making process when evaluating potential shunt failure.