Parental diagnosis of shunt malfunction in children

Introduction

Evaluation of shunt for malfunction and blockage is a common neurosurgical clinical scenario in day-to-day paediatric neurosurgery practice. Parental concern for shunt malfunction is normally considered a reliable indication for undertaking further assessment of a child with a shunt. Accuracy of parental diagnosis of shunt malfunction likely is dependent on combination of parental knowledge and patient symptomatology and familiarity of symptoms in relation to previous shunt malfunction symptoms. There are currently no UK studies on accuracy of parental diagnosis of shunt malfunction. We undertook this study to investigate sensitivity and specificity of parental diagnosis in our tertiary paediatric neurosurgical unit to identify any factors that can be used for further education and raising awareness in parents with children with shunts.

Methods

We undertook a review of all referrals with suspected shunt malfunction to our tertiary paediatric neurosurgical unit over a period of 10 months. All referrals and presentations were evaluated for parental concern for likelihood of shunt malfunction and marked as yes or no. Further information gathered included demographics, age, sex, symptoms at presentation, previous history of shunt revision, cases of shunt malfunction and part of the shunt revised. Sensitivity, specificity, positive and negative predictive values and accuracy were then calculated and factors associated with positive diagnosis of shunt malfunction analysed.

Results

Overall, 100 referrals with suspected shunt malfunction were made over 10 months period. Mean age for children at the time of referral was 6.3 years with an M/F ratio of 42M:58F. Twenty-one shunt malfunction cases were identified leading to overall revision rate of 21%. Parental evaluation of shunt function had sensitivity of 90.4%, specificity of 10.1%, positive predictive value of 21.1% and negative predictive value of 80%. Drowsiness and higher number of symptoms at presentation were associated with true positive diagnosis of shunt malfunction. No link was identified with true diagnosis of shunt malfunction with other symptoms of shunt malfunction or previous history of shunt revision and age of the patient.

Conclusion

Parental diagnosis of shunt malfunction has high sensitivity and negative predictive value and low specificity and positive predictive value. Increased number of symptoms as well as drowsiness were associated with correct parental diagnosis of shunt malfunction. Educational programs and parental training can be valuable in increasing awareness about shunt malfunction signs.