Postsurgical infection after myelomeningocele repair: a multivariate analysis of 60 consecutive cases.

Objective: We investigate the demographic, clinical and surgical variables associated with wound and ventriculoperitoneal (VP) shunt infections in a well-defined group of patients submitted to neurosurgical myelomeningocele repair.

Methods: We analyzed the data of sixty consecutive patients with a myelomeningocele diagnosis submitted to neurosurgical repair between January 2002 and December 2005. Multiple logistic regression analysis identified clinical, demographic and neurosurgical variables that were independently associated with the occurrence of wound and VP shunt infections.

Results: Seven patients (11.7%) developed wound infections after myelomeningocele repair and two (3.3%) presented with sepsis unrelated to the neurosurgical procedures. Forty-six patients (76.7%) received a VP shunt and nine of them (19.6%) had VP shunt infection. There was a non-significant trend (p=0.09) for a higher association between thoracic than lumbar or sacral topography and the occurrence of any type of infection. Among patients who underwent VP shunt placement, there was a non-significant trend for a higher association between VP shunt infection and thoracic topography compared to lumbar or sacral regions (adjusted OR 4.3; CI 95% 0.7-24.7; p=0.10). Evans' index scores higher than 70 were ten times more associated with VP shunt infection (adjusted OR 10.5; CI 95% 1.6-67.4; p=0.01) than lower scores.

Conclusion: The thoracic topography of myelomeningocele has a trend for a higher association with infection in general and VP shunt infection. Evans' index scores higher than 70 were independently associated with VP shunt infection.