The main indications for supratentorial operation were intracranial space occupying lesions, especially tumors. Resection of space occupying lesions often require large dural openings, which can be result in a higher risk of reconstruction failure with associated CSF leakage than smaller defects.[5] The incidence rate of CSF leakage in the elective supratentorial procedures varied from 1.2–10.9%. [1, 10, 37, 2, 11] There are no precise data on the incidence rate of CSF leakage because it can be described in a variety of ways.
Specific risk factors for post-craniotomy CSF leakage include recurrent operation, the state of the dura mater, the duration of the surgery, the size of the dural opening, or serious comorbidities.[5] CSF leakage leads to surgical site infection, meningitis, wound healing impairment, and hydrocephalus.[5, 14, 8] Furthermore, CSF leakage and common complications result in increased morbidity, hospital stay lengthening, surgical revision, and enhanced expenditures.[11]
Recently, several reports have shown that a non-WTDC does not always lead to a higher incidence of CSF leak. Barth et al.[2] intraoperatively randomized enrolled patients who were experiencing supratentorial procedures in which primary watertight dural repair could not be accomplished into 2 groups (secondary WTDC group and adaptive dural closure group). In the secondary WTDC group, watertight suture was reached using grafts, while in the adaptive non-WTDC group, the dura mater was sutured in interrupted simple sutures placed 2–3 cm apart. Between the groups, there was no significant difference in the incidence of postoperative CSF leak and infection. However, in adaptive dural closure group, surgeries needed less time and costs for dural closure. Alwadei et al.[1] divided the patients who had undergone supratentorial craniotomies into 2 groups (dura closed group and open group), according to surgeon preference rather than the intraoperative conditions. No significant differences were found between 2 groups in terms of CSF leak, infection, and surgical site swelling. Sade et al.[37] also found that in intracranial meningioma surgery, watertight dural closure was not better than dural reconstruction using the onlay dura graft according to postoperative complications.
In our study, the WTDC group had a rate of 9.7% in subgaleal fluid collection, while the non-WTDC group had a rate of 11.3%, although this difference was not significant (p = 0.502). Our results are consistent with those in previous literatures. A possible explanation for CSF leak in WTDC patients might be that acute increased postoperative intracranial pressure could widen the gap of sutures and enlarge the pinhole, due to extubation, postoperative headache, cough and inappropriately moving the patients. Even the pinholes produced by surgical needles have been shown to prevent a “watertight” dural repair.[24, 29]
As mentioned in the literature review, the incidence rate of postoperative infection in elective cranial surgeries ranged from 0.9–8.7%.[15, 1, 42, 37, 45, 19] These 2 complications, CSF leakage and infection, can also be dependent variables, so that infection is sometimes supposed to cause CSF leakage[41], and vice versa[28]. Despite limited comparability, frequency of postoperative infection in our series was generally consistent with findings in previous series.[15, 36, 45, 2] The most important clinically relevant finding was that watertight dural repair could decrease the risk infection in supratentorial procedures. This finding is contrary to previous studies which have suggested that a watertight dural repair is not essential in supratentorial craniotomies.[1, 42, 37, 2]
A possible explanation for this might be that all the enrolled patients were diagnosed as space-occupying lesions, which require large dural openings and involve a variety of surrounding tissues. Without watertight dural closure, CSF could circulate between the epidural and subdural space and contact skull, galea, muscle, and scalp.[37] The whole process could increase the probability of infection. Although subcutaneous fluid was actively treated, the circulation of CSF has already completed. With watertight dural suture, CSF leak did not progress when the increased postoperative intracranial pressure disappeared. After extraction of epidural CSF, the circulation of CSF between the epidural and subdural space was absent.
There is no clear definition of post-craniotomy headaches (PCHs) currently. Several reports have suggested the following diagnostic criteria of PCHs: (1) with pulsatile or pounding pain in any part of the head; (2) with variable degree that arises postoperatively and having a 7–14 days duration; (3) with negative radiographic and clinical manifestations related with increased intracranial pressure.[1, 35, 13] There is some evidence that the incidence of PCHs may vary according to the surgical technique.[38, 22, 39] However, very little was found in the literature on the question of the relationship of PCHs and WTDC in supratentorial craniotomy. Besides the above post-craniotomy complications, Alwadei et al.[1] also paid attention to post-craniotomy headaches. Significantly greater association of PCHs was presented in the dura closed group (p = 0.001). However, the finding of the current study does not support the previous research. With 423 patients in the WTDC group, and 275 in the non-WTDC group, they suffered from a greater incidence of post-craniotomy headaches in the WTDC group (13.5% vs.9.5% in the non-WTDC group), but without statistical significance (p = 0.109). It can therefore be assumed that watertight dural suture were not directly associated with post-craniotomy headache.
There were some limitations to our study. At first, being limited to the retrospective nature, this monocentric, non-randomized study has the inherent risks of bias. Then, 95.4% enrolled patients in our study developed tumors which often require large dural openings. Considerably more work will need to be done to determine whether the watertight dural closure can be applied to other diseases, like a vascular lesion, in supratentorial craniotomy. Finally, further research should be undertaken to investigate the dural suture and its relationship to the occurrence of CSF leak and postoperative infection.