The Prediction of Shunt Response in Idiopathic Normal-Pressure Hydrocephalus Based on Intracranial Pressure Monitoring and Lumbar Infusion.

BACKGROUND

Intracranial pressure (ICP) monitoring and infusion studies have long been used in the preoperative workup of patients with suspected idiopathic normal-pressure hydrocephalus (iNPH). We have analysed the predictive values of different measures derived from both investigations, emphasising the differences between responders and nonresponders.

MATERIALS AND

METHODS

ICP monitoring and lumbar infusion studies were routinely performed during a 6-year period. Shunting was proposed when the resistance to cerebrospinal fluid outflow (ROUT) >12 mmHg/ml/min and/or a minimum 15 % of slow waves were detected. The outcome was evaluated 6 months after surgery. Recorded data from ICP monitoring were mean pressure and pulse amplitude, the total percentage of slow waves and the presence of different types of slow waves following the classification proposed by Raftopoulos et al. Recorded data from lumbar infusion studies were mean values of pressure and pulse amplitude during three epochs (basal, early infusion and plateau), ROUT and the pulsatility response to the increase in mean pressure during the infusion. This response was quantified by two pulse amplitude indexes: the pulse amplitude index during the early infusion stage (A1) and the pulse amplitude index during the plateau stage (A2).

RESULTS

Thirty shunted patients were evaluated at the end of the follow-up and 23 (76.7 %) of them improved. Differences in the percentage of slow waves, ROUT and both pulsatility indexes were not statistically significant. The proportion of patients with great symmetrical waves and pulse amplitude during the early infusion stage were higher in responders (p < 0.05). The predictive analysis yielded the highest accuracy, with ROUT and A1 as a logical "OR" combination.

CONCLUSION

The combined use of ICP monitoring and lumbar infusion to forecast the response to shunting in patients with suspected iNPH did not improve the accuracy provided by any of them alone.