Dural And Skull Base Reconstruction To Eliminate Postoperative Complications Related To Cerebrospinal Fluid Leakage: Technical Nuances And Surgical Outcomes

DOI: https://doi.org/10.21203/rs.3.rs-636279/v1

Abstract

BACKGROUND: Cerebrospinal fluid (CSF) leakage is a common complication associated with neurosurgical procedures, and eliminating postoperative CSF leaks remains challenging.

OBJECTIVE: Unique applications of free or vascularized flaps for watertight dural and skull base reconstruction are presented.

METHODS: A total of 512 cases of cranial surgeries were examined focused on the postoperative complications that required revision surgery. These cases were divided into skull base (SB) lesions and non-SB lesions according to the site of dural opening. The postoperative complications were analyzed according to their relationship to the CSF. A free flap, including fascial, myofascial, or pericranial flap, is routinely prepared for later use in dural closure in all cranial surgeries. A pedicled vascularized pericranial flap (VPF) is also used for SB dural reconstruction in extended surgical procedures.

RESULTS: There were 434 cases of SB lesions and 78 cases of non-SB lesions. Of the 512 surgical cases, 27 (5.3%) required revision surgery due to patients’ postoperative conditions or complications. There were 9 cases (1.8%) of CSF-related and 18 cases (3.5%) of non-CSF-related complications. Eight of nine cases of CSF-related complications (88.9%) were seen in posterior fossa surgery. Postoperative CSF leaks that required revision surgery were seen in 4 cases (0.8%). SB reconstruction using a VPF was performed in 40 cases of SB surgeries. No postoperative infection was seen in these 40 cases, and only 1 case (2.5%) of pseudomeningocele, which required revision surgery, occurred.

CONCLUSION: Dural reconstruction using a fascial flap or VPF plays an important role in achieving successful surgery. 

Introduction

Postoperative cerebrospinal fluid (CSF) leaks or subcutaneous fluid collections (pseudomeningocele) are among the most common complications in neurosurgery, especially in skull base (SB) surgery. A CSF leak may cause deterioration of a patient’s activity, a prolonged hospital stay, and surgical site or intracranial infection. To prevent postoperative CSF leaks or pseudomeningocele, watertight dural closure is absolutely essential. We have used a fascial graft for secured dural closure and reconstruction. In any SB case that requires an extended or invasive SB approach, reconstruction of the SB using a vascularized pericranial flap (VPF) is needed. Our unique use of a fascial flap or VPF in neurosurgical interventions is presented, and the surgical outcomes are analyzed.

Material And Methods

Patients’ characteristics

A total of 512 cases of intracranial surgery, including tumor, vascular, traumatic, and functional surgeries, operated by the first author from 2008–2018 were examined, focusing on the postoperative complications that required revision surgery. These postoperative complications were divided into CSF-related and non-CSF-related. CSF leakage was characterized by rhinorrhea, otorrhea, and leakage from the operative wound. Patients were divided into those with SB lesions and those with non-SB lesions according to the site of dural opening. The SB lesions included anterior SB, middle SB or infratemporal fossa (ITF) lesions, and posterior SB or craniovertebral junction (CVJ) lesions. Postoperative CSF-related complications, which did not respond to conservative care, required additional surgical manipulation, including revision surgery for dural repair or plasty, wound opening, debridement or irrigation. Conservative procedures such as continuous lumbar drainage placement with pressure dressing on the surgical site were not included as revision surgeries. The Institutional Review Board for Clinical Research approved this retrospective study. All patients had previously given informed consent for the surgery and all diagnostic and therapeutic interventions/techniques applied; the Review Board ruled that additional patient consent for inclusion into this retrospective study was not required.

Fascial graft (free flap)

We routinely prepare a fascial, myofascial, or pericranial flap, which is harvested at the surgical site, for later use in watertight dural closure in all cranial surgeries, especially in posterior fossa surgery. The skin incision is made through the subcutaneous fat layer, and then the skin is undermined widely to expose the surface of the musculofascial layer (Video 1). An incision is made through the musculofascial layer to elevate a thin musculofascial flap without any laceration from the muscle layer [9]. This flap is used as an autologous graft for dural closure. An inlaid flap, which is larger than the dural opening in size, is applied to cover all needle holes of suturing from inside of the dura (hole-cover technique) (Fig. 1). The inlaid flap reduces small CSF leaks from the needle holes by tight contact with the dura. A leak test using the Valsalva maneuver and physiological saline water infusion is performed to confirm watertight closure. Dural closure is reinforced by using fibrin glue (Bolheal®) (Teijin, Tokyo, Japan) and Surgicel® (Ethicon US, LLC, Cincinnati, OH, USA), followed by bone flap fixation by a titanium plate and screw.

Vascularized pericranial flap (pedicled flap)

A VPF is used for the SB dural reconstruction mainly in extended and invasive surgical procedures. The scalp is elevated by sharply dissecting the true galea off from the subgaleal loose areolar connective tissue, which lies over the pericranium (Fig. 2, Video 2). The non-flap-side scalp edge is undermined, and the pericranium is elevated from the undermined deep end. The VPF and the fascia are sharply elevated as a large flap is separated from the temporal muscle and the sternocleidomastoid muscle (SCM). A thin portion of the SCM and the temporal muscle may be elevated together with the fascia to maintain continuity with the pericranium and for good vascularity. A wide VPF can be cut freely to provide vascular tissue to secure the closure of the dural incision or to cover the opened air sinuses, or to fill a dead space due to a bony defect or large tumor cavity.

Skull base reconstruction

When the SB dura mater is partially removed with a tumor, an appropriate size of abdominal fascia is applied for the primary dural closure. Then, the SB side of the fascia is fixed onto the bone surface of the SB at some points by one or two titanium microplates and screws (microplate-bridge technique) [18] (Fig. 3, Video 2). The dural side of the fascia can then be sutured easily to the dura watertightly. Some abdominal fat grafts cut into multiple narrow strips are placed on the edge of the SB side fascia in multilayered fashion with fibrin glue to compress the gap between the SB bone and the fascia. The VPF is brought to the extradural space through an incision made on the temporal muscle and then sutured on the dura widely to cover the reconstructed area. Additional fat grafts are applied onto the sutured VPF to eliminate any dead spaces.

Results

Type of surgeries

There were 434 cases of SB lesions and 78 cases of non-SB lesions (Table 1). The SB lesions included anterior SB lesions in 109 cases, middle SB or ITF lesions in 19 cases, and posterior SB or CVJ lesions in 306 cases. There were 302 cases of tumor surgery, 104 cases of functional surgery, 95 cases of vascular surgery, and 10 cases of trauma surgery. Tumor surgery included 133 cases of meningiomas, 91 cases of schwannomas, and others. Functional surgery included 102 cases of microvascular decompression for trigeminal neuralgia and hemifacial spasm. Vascular surgery included 79 cases of aneurysmal neck clipping and others. Trauma surgery included 8 cases of acute epi- and subdural hematomas, 2 cases of skull fracture, and others.

Table 1. Patients’ characteristics  

CVJ; craniovertebral junction, ITF; infratemporal fossa, SB; skull base

Postoperative complication and revision surgery

Of the 512 surgical cases, 27 (5.3%) required revision surgery due to the patient’s postoperative condition or complications (Fig. 4). There were 9 cases (1.8%) of surgical site infection (SSI), 5 cases (1%) of pseudomeningocele, 5 cases of intracranial hemorrhage, and others (Table 2). These complications were divided into CSF-related and non-CSF-related complications. The 9 cases (1.8%) of CSF-related complications included 5 cases of pseudomeningocele and 4 cases of CSF leaks requiring revision surgery. Eight of 9 cases (88.9%) of CSF-related complications were seen in posterior fossa surgery. The overall rate of CSF-related complications in posterior fossa surgery in the present series was 2.6% (8/306). Non-CSF-related complications were seen in 7 cases in the posterior SB, 5 in the non-SB, 2 cases in the anterior SB, and 1 in the middle SB.

Table 2. Postoperative complications that require revision surgery of the 512 cases 

CSF; cerebrospinal fluid, SSI; surgical site infection. * A patient who presented with a CSF leak from the wound developed to the SSI.

SB reconstruction using a VPF was performed in 30 cases of posterior SB, 7 cases of middle SB or ITF, and 3 cases of anterior SB surgeries. No postoperative infection was seen in these 40 cases, and only 1 case (2.5%) of pseudomeningocele occurred that required revision surgery.

Infection

Nine cases (1.8%) of SSI, which did not respond to antibiotics, required revision surgery. SSI was divided into CSF-related infection (1 case) and non-CSF-related infection (8 cases). All cases required removal of the bone flap and irrigation or wound debridement if necessary. The CSF-related infection occurred in 1 case subsequent to a CSF leak. Non-CSF-related infections were seen in 4 cases of posterior SB, 3 cases of non-SB, and 1 case of anterior SB.

CSF leak

Four cases (4/512; 0.8%) of CSF leak, which did not improve by lumbar drainage (rhinorrhea in 2, wound leak in 2) required revision surgery. All CSF leaks were seen in posterior SB surgeries (4/306; 1.3%). Two cases of rhinorrhea occurred in the retrosigmoid approach, 2 cases of wound leak were seen in the combined retrolabyrinthine and retrosigmoid approach, and the midline suboccipital approach.

Pseudomeningocele

Some pseudomeningoceles regressed spontaneously with a pressure dressing with or without lumbar drainage. Revision surgery was required in 5 cases (5/512; 1%) of pseudomeningoceles (posterior SB 4, non-SB 1). Dural plasty using an additional fascial graft and re-suturing of a loosened stitch were performed in three cases of pseudomeningocele, which occurred in the posterior fossa. A ventriculo-peritoneal shunt was done for an intractable pseudomeningocele, which was seen in the case of a huge calvarial meningioma with a wide-ranging skull bone and dural defect.

Discussion

Conceptually, a watertight dural closure is a technically simple proposition. However, postoperative CSF leakage has been the most frequent serious complication, especially in posterior fossa surgery, because of its gravity-dependent location [1–7,13,23,26–28,30,32]. Many reports have discussed factors leading to postoperative CSF leaks after resection of vestibular schwannomas, which is notorious for postoperative CSF leaks [3,22,28,31,32]. They indicated hydrocephalus, tumor size, translabyrinthine approach, or longer operative times were risk factors for CSF leaks following vestibular schwannoma surgery. However, some authors disagreed [13,25].

Previously, we reported our surgical outcomes and complications of a consecutive series of 410 cases of vestibular schwannomas [21]. There were significant differences in the incidence of CSF leaks by surgeon. When dural closure was performed by residents, the incidence of CSF leaks was 13.9%. However, when closure was performed by faculty or experienced surgeons, the rate of CSF leaks decreased to 5.5%. In the present series, 2 of 4 cases (50%) of CSF leaks and 2 of 5 cases (40%) of pseudomeningoceles occurred after dural closure by residents. Although 8 of 9 cases (88.9%) of CSF-related complications were seen in posterior SB surgery, their overall incidence in the 306 cases of posterior SB surgery was 2.6%. Furthermore, all CSF leaks were seen in posterior SB surgeries (4/306; 1.3%), which is relatively lower than in past reports (6.6%-13%) [4, 5, 8].

Throughout the surgery, the reflected dural flap dries up and shrinks easily by coagulation for hemostasis or heat from the illumination of the microscope. Moreover, the posterior fossa dura is generally thin and friable. In such situations, to pull back the shrunken dural flap to the original position for closure is not easy, and the putting of a needle on it or tying suture easily causes dural tears, which makes a watertight closure impossible. Closure using a fascial flap reduces the tension on the shrunken dura. An inlaid fascial flap, which is incised larger than the dural incision, will cover the stitch hole from inside. Moreover, the fascial flap will be compressed to the inner surface of the dura by the pressure of the CSF, which will make dural closure tighter. Furthermore, an autologous tissue is easy to stick on to the dura mater.

We routinely use a two-inch, lazy-S-shaped skin incision in the retrosigmoid approach. A pressure dressing, which may contribute to reducing dead space to create a pseudomeningocele or a direct conduit for CSF to reach the skin, is applied postoperatively for 72 hours for all retrosigmoid cases. Interestingly, all pseudomeningoceles occurred in cases in which the skin incision was larger than two inches (three-inch, postauricular C-shaped incision in 1, three-inch, lazy-S incision in 2, four-inch linear incision in 1, a large rectangular incision in 1). These results indicate that a wide range of subperiosteal detachment from the bone with the existence of a CSF leak from the dural gap may cause the formation of a postoperative pseudomeningocele.

Vascularity is a critical element for adherence between the tissues in the process of wound healing and for prevention of infection. Reconstruction with vascularized autologous tissue is preferred where watertight closure of the dura is required, and significant potential for infection exists. It has been well established to use a vascularized tissue flap for the repair of large SB defects [10–12,14–17,19,20,22,29]. In other words, aggressive surgical approaches for the treatment of SB tumors are warranted with the availability of reconstructive techniques using the VPF, which minimize postoperative morbidity [29]. In cases of a recurrent situation, where a flap has already been used, the transfer of a VPF from an untouched area remains the ultimate procedure. For this reason, in cases of re-do surgery for SB tumor recurrence, a skin incision much larger than the previous one should be made to effectively obtain the VPF (Fig. 5).

There are several limitations to this study. The first is that this study included only a few cases of malignant neoplasms of the anterior SB or nasofrontal region, or extensive traumatic fractures that involve the paranasal sinuses, which may result in large bone defects with the potential to lead to much higher complication rates than other regions. Therefore, the lower rate of postoperative infection in this study cannot be compared simply with that of other studies of anterior SB reconstruction [11,12,14,24,29]. Second, the distribution of the lesions was not equal, and the types of surgery varied.

Conclusion

We believe that the incidence of CSF-related complications is simply a reflection of the surgeon’s skill, attention, experience, and the way of reconstructing the dura mater. The present surgical outcomes demonstrated that dural reconstruction using a fascial flap and SB reconstruction using a VPF would significantly reduce the risk of postoperative CSF leakage and infection. Therefore, synthetic materials, as potential dural substitutes, should only be used when the autologous grafts are not available.

Abbreviations

CSF, cerebrospinal fluid; CVJ, craniovertebral junction; ITF, infratemporal fossa; SB, skull base; SCM, sternocleidomastoid muscle; SSI, surgical site infection; VPF, vascularized pericranial flap

Declarations

Funding: Not applicable

Conflict of Interest: Not applicable

Availability of data and material: Not applicable

Code availability: Not applicable

Authors’ contributions: Not applicable 

Ethics approval

The Institutional Review Board for Clinical Research of Tokai University Hospital approved this retrospective study (approval No. 19R218). 

Consent to participate: Not applicable 

Consent for publication: Not applicable 

Acknowledgements

The authors would like to thank Hikaru Maruiwa, MD, Kenji Sugita, MD, Nobuyuki Kaneko, MD, and Kazumi Ohmori, MD, for providing the follow-up studies (Figure 5A-D) and a cadaveric photo (Figure 2C). 

Corresponding Author

Yoichi Nonaka

143 Shimokasuya, Isehara, Kanagawa 259-1193, Japan

TEL: +81 463 (93) 1121, FAX: +81 463 (93) 1295

E-mail address: [email protected]

ORCID ID: 0000-0001-5509-6175

References

1. Bani A, Gilsbach JM (2002) Incidence of cerebrospinal fluid leak after microsurgical removal of vestibular schwannomas. Acta Neurochir (Wien) 144:979-982

2. Black P (2000) Cerebrospinal fluid leaks following spinal or posterior fossa surgery: use of fat grafts for prevention and repair. Neurosurg Focus 9:e4

3. Brennan JW, Rowed DW, Nedzelki JM, Chen JM (2001) Cerebrospinal fluid leak after acoustic neuroma surgery: influence of tumor size and surgical approach on incidence and response to treatment. J Neurosurg. 94:217-223

4. Copeland WR, Mallory GW, Neff BA, Driscoll CL, Link MJ (2015) Are there modifiable risk factors to prevent a cerebrospinal fluid leak following vestibular Schwannoma surgery? J Neurosurg 122:312-316

5. Dubey A, Sung WS, Shaya M, Patwardhan R, Willis B, Smith D, Nanda A (2009) Complications of posterior cranial fossa surgery-an institutional experience of 500 patients. Surg Neurol 72:369-375

6. Falcioni M, Mulder JJ, Taibah A, De Donato G, Sanna (1999) No cerebrospinal fluid leaks in translabyrinthine vestibular schwannoma removal: reappraisal of 200 consecutive patients. Am J Otol 20:660-666

7. Felbaum DR, Mueller K, Anaizi A, Marson R, Jean WC, Voyadzis JM (2016) Preservation of the myofascial cuff during posterior fossa surgery to reduce the rate of pseudomeningocele formation and cerebrospinal fluid leak: A technical note. Cureus 8:e946.DOI 10.7759/cureus.946.

8. Fishman AJ, Marrinan MS, Golfinos JG, Cohen NL, Roland JT Jr (2004) Prevention and management of cerebrospinal fluid leak following vestibular schwannoma surgery. Laryngoscope 114:501-505

9. Fukushima T, Nonaka Y (2010) Combined Neuro-otology & Neurosurgery Skull Base Approaches to the Temporal bone. Fukushima T, Nonaka Y (eds); Fukushima Manual of Skull Base Dissection, 3rd edn. AF-Neurovideo Inc, Raleigh, NC; pp 210-249.

10. Guinto G, Guinto Y (2015) Reconstruction techniques in skull base surgery. World Neurosurg 83:17-18

11. Harvey RJ, Parmar P, Sacks R, Zanation AM (2012) Endoscopic skull base reconstruction of large dural defects: A systematic review of published evidence. Laryngoscope 122:452-459

12. Hechem RA, Elkhatib A, Beer-Furlan A, Prevedello D, Carrau R (2016) Reconstructive techniques in skull base surgery after resection of malignant lesions: a wide array of choices. Curr Opin Otolaryngol Head Neck Surg 24:91-97

13. Hoffman RA (1994) Cerebrospinal fluid leak following acoustic neuroma removal. Laryngoscope 104:40-58

14. Hoffman TK, Hindy NE, Muller OM, Schuler PJ, Bergmann C, Hierner R, et al (2013)Vascularised local and free flaps in anterior skull base reconstruction. Eur Arch Otorhinolaryngol 270:899-907

15. Ito E, Watanabe T, Sato T, Ichikawa M, Oda K, Matsumoto Y, et al (2012) Skull base reconstruction using various types of galeal flaps. Acta Neurochir 154:179-185

16. Jalisi S, O’Gara B, Toshkezi G, Chin L (2015) Local vascularized flap reconstruction of the skull base-clinical outcomes and analysis. World Neurosurg 83:87-92

17. Komatsu F, Imai M, Hirayama A, Hayashi N, Oda S, Shimoda M, Matsumae M (2017) Endoscopic middle cranial fossa reconstruction with a subtemporal keyhole. World Neurosurg 108:157-162

18. Kusumi M, Fukushima T, Aliabadi H, Mehta A, Noro S, Rosen C, Fujii K (2012) Microplate-bridge technique for watertight dural closures in the combined petrosal approach. Neurosurgery 70:264-269

19. Kusumi M, Fukushima T, Mehta AI, Cunningham CD 3rd, Friedma AH, Fujii K (2013) Middle fossa approach for total resection of petrous apex cholesterol granulomas: use of vascularized galeofascial flap preventing recurrence. Neurosurgery. 72(1 Supple Operative):77-86

20. Macia G, Picon M, Nunez J, Almeida F, Alvarez I, Acero J (2015) The use of free flaps in skull base reconstruction. Int J Oral Maxillofac Surg 45:158-62

21. Nonaka Y, Fukushima T, Watanabe K, Friedman AH, Sampson JH, McElveen JT, Cunningham CD III, Zomorodi AR (2013) Contemporary surgical management of vestibular schwannomas; Analysis of complications and lessons learned over the past decade. Neurosurgery. 72(2 Suppl Operative):ons103-115

22. Pirouzmand F, Tator CH, Rutka J (2001) Management of hydrocephalus associated with vestibular schwannoma and other cerebellopontine angle tumors. Neurosurgery 48:1246-1254.

23. Pulec JL (1994) Technique to avoid cerebrospinal fluid otorhinorrhea with translabyrinthine removal of acoustic neuroma. Laryngoscope 104:382-386

24. Safavi-Abbasi S, Komune N, Archer JB, et al (2016) Surgical anatomy and utility of pedicled vascularized tissue flaps for multilayered repair of skull base defects. J Neurosurg 125:419-430 25. Sanna M, Taibah A, Russo A, Falcioni M, Agarwal M (2004) Perioperative complications in acoustic neuroma (vestibular schwannoma) surgery. Otol Neurotol 25:379-386

26. Selesnick SH, Liu JC, Jen A, Newman J (2004) The incidence of cerebrospinal fluid leak after vestibular schwannoma surgery. Otol Neurotol 25:387-393

27. Sen A, Green KM, Khan MI, Saeed SR, Ramsden RT (2006) Cerebrospinal fluid leak rate after the use of BioGlue in translabyrinthine vestibular schwannoma surgery: a prospective study. Otol Neurotol 27:102-105

28. Slattery WH III, Francis S, House KC (2001) Perioperative morbidity of acoustic neuroma surgery. Otol  Neurotol 22:895-902

29. Snyderman CH, Janecka IP, Sekhar LN, Sen CN, Eibling DE (1990) Anterior cranial base reconstruction: Role of galeal and pericranial flaps. Laryngoscope 100:607-614

30. Stieglitz LH, Giordano M, Gerganov V, et al (2010) Petrous bone pneumatization is a risk factor for cerebrospinal fluid fistula following vestibular schwannoma surgery. Neurosurgery 67 (2 Suppl Operative):509-515

31. Sughrue ME, Yang I, Aranda D, et al (2011) Beyond audiofacial morbidity after vestibular schwannoma surgery. J Neurosurg 114:367-374

32. Yasargil MG (1996) Management of acoustic neuromas, in Microsurgery. Vol IVB. Thieme-Verlag, New York, pp 100-123.