Abstract
Summary Nineteen publications were reviewed and subjected to a combined statistical analysis (meta-analysis) regarding the influence of study design factors upon reported headache and total symptom incidences after lumbar iohexol myelography. A significant association was found between reported side effects on one hand and needle diameter, follow up time and the method of questioning respectively on the other. The combination of long follow-up time and specific questioning and the combination between larger diameter (20G) needles and long follow-up time, both seemed to be strong predictors for reporting high side effect incidences. Nine studies were similarly analyzed regarding the influence of early ambulation and contrast type upon reported headache incidences. Early ambulation significantly increased headache after iohexol or iopamidol lumbar myelography as opposed to metrizamide myelography.