Neurosurgery. 2026;99(4):873-881
No universally optimal sequence emerges when patients require both cranioplasty and permanent CSF diversion after decompressive craniectomy. Holmgren and colleagues found broadly similar unadjusted outcomes, but an adjusted analysis associated shunting first with fewer cranioplasty removals. This potentially important signal must be interpreted alongside substantial differences in why and when each strategy was chosen.
Study and key findings
This retrospective study from four Swedish centers included 99 patients: 37 underwent cranioplasty first, 37 simultaneous surgery and 25 shunting first. Median follow-up after cranioplasty was 58 months.
Cranioplasty removal occurred in 30%, 24% and 8%, respectively; the unadjusted comparison was not statistically significant (P=.121). Shunt revision rates were 24%, 35% and 36%, also without a significant difference. Functional outcomes did not differ significantly.
After adjustment for cranioplasty material and the interval from decompressive craniectomy to reconstruction, shunting first was associated with lower odds of cranioplasty removal than cranioplasty first: OR 0.12, 95% CI 0.01-0.64, P=.031. Only two removals occurred in the shunt-first group, making this estimate sensitive to small event numbers.
Interpretation and limitations
Treatment groups were not clinically interchangeable. External cerebral herniation prompted shunting in none of the cranioplasty-first patients, compared with 70% of simultaneous and 44% of shunt-first patients. Center, implant material and timing also differed. Residual confounding therefore prevents a causal interpretation of the adjusted association.
The absence of significant differences in other outcomes does not demonstrate equivalence. Because every included patient ultimately received both procedures, the study cannot establish whether cranioplasty first reduces the need for permanent shunting. Detailed valve-management data were unavailable, so no specific pressure-setting protocol can be derived.
Clinical takeaway
Sequence should remain individualized to the CSF disorder, cerebral contour and reconstructive requirements. The association favoring shunting first for cranioplasty retention merits further investigation, but does not justify a universal shunt-first policy or establish identical safety for simultaneous surgery.