It is common practice to reapproximate the dura to prevent the leakage of cerebrospinal fluid (CSF) at the end of craniotomy procedures. Dural closure occurs by reapproximation of the dural edges with suture or by inserting a graft material between the dural defect. CSF leakage is a latent risk of cranial surgery with dural opening. According to the literature, CSF leaks vary from 4% in transsphenoidal procedures to 32% in posterior fossa procedures8). Recognizing the risk factors for this complication and improving the technique of dural closure may reduce the associated morbidity and its surgical burden.

Numerous techniques and materials have been described for dural closure. Most techniques utilize autologous tissue grafts of fat, muscle or pericranium, with or without the use of postoperative lumbar CSF drainage15,17). Various adhesive substances for duraplasty are currently in use, including collagen matrix and fibrin glue10,11). However, dural closure with synthetic grafts has been suggested to contribute to the occurence of infection and CSF leak9,14). Therefore, surgeon should decide whether to use synthetic dural substitute in conflict between the potential benefit of improved dural closure with the potential increase in infection rate.

We have used Duraform as the final layer of dural closure in elective craniotomy since 2013. The purpose of this study was to evaluate the usefulness of Duraform(Codman & Shurtleff Inc., Raynham, MA, USA) as dural sealant based on the authors’ experience over the past 3 years. We compared our standard dural closure by continuous suture alone to the same closure with the addition of Duraform on top of the suture. Furthermore, we assessed the risk factors for CSF leakage and potentially related complications after craniotomy.

Three hundred sixty-three patients were included with mean follow-up of 13±standard deviation (SD) of 9.2 months in the study group (n=117) and 18±11.2 months in the control group (n=246). Patient demographic and clinical characteristics are shown in Table 1. There were no statistically significant differences between groups.

As shown in Table 2, postoperative CSF leaks were identified in 6 patients (5.1%) in the Duraform application group and in 31 (12.6%) in the control group. In the study group, only two (1.7%) patients developed meningitis, and no other complications were reported during the follow-up period. However, in the control group, 19 patients (7.7%) developed meningitis during the postoperative period. The mean postoperative hospital stay was quite shorter in the study group (9.1±1.3 days) than in the control group (12.6±2.8 days).

Univariate logistic regression analysis identified diabetes mellitus, use of dural patch, and infratentorial operative location as having a significant association with postoperative CSF leak. Similarly, univariate predictors of CSF leak were found to be diabetes mellitus, length of operation, estimated blood loss, and infratentorial operative location (Table 3).

Multivariate logistic regression modeling identified infratentorial operative location (odds ratio [OR] 4.68, 95% confidence interval [CI] 1.34-15.98; p=0.019) as significant predictors of CSF leak (Table 4). No factors found to be predictor of infection in multivariate analysis (Table 5).

The result of our study indicates that the addition of Duraform on top of the dural suture during elective craniotomy can be useful in preventing CSF leak after cranial surgery and obviates the need for interventions due to CSF collection, which can shorten the length of hospital stay. Several studies have described the use of sealants as helpful in preventing CSF leakage in supratentorial1,13), infratentorial3), skull base12), and spinal6,8) surgeries which consistent with our study. However, instead of getting the benefit of improved dural closure, the surgeons are faced with the potential increase in infection associated with the implantation of an artificial material5). Any foreign material implanted into the body stimulates a host response such as inflammatory cell infiltration or neovascularization2,7). In contrary to prior reports, we observed a lower frequency of serious complications, such as meningitis and epidural empyema, in the Duraform application group. Hutter et al.5) suggested a theory for explanation of this finding that optimal sealing of the intradural compartment by dural augmentation avoids bacterial migration through microlesions even by the dural suture itself. Furthermore, our analysis suggests that patient-centered variables are more influential in the development of infection than the selection of dural substitute materials or dural repair techniques.

Among the factors examined, infratentorial craniotomy was found to be associated with postoperative CSF leak. Infratentorial surgery predisposes a patient to CSF leak because the hydrostatic pressure of the fluid column puts additional stress on the dural suture16). The association between CSF leak and diabetes mellitus has so far rarely been identified in the current literature. Possible explanation for this finding might be that diabetes mellitus impairs wound healing resulting in failure of dural closure.

Several limitations of the present study include its retrospective nature, single center experience, and its relatively small cohort. Additionally, the results are not generalizable at different institutions due to different patient management protocols. Further studies encompassing a larger cohort of patients need to be established.

The present study shows that addition of dural sealant upon the dural suture reduces postoperative CSF leak and surgical site infection after elective craniotomy with dural opening. Its clinical use for dural augmentation was safe and not related to adverse events. Furthermore, clinical risk factors exist for postcraniotomy CSF leak and infection: diabetes mellitus and infratentorial operative location.