Journal of Clinical Neuroscience xxx (2018) xxx–xxx

Contents lists available at ScienceDirect

Journal of Clinical Neuroscience

journal homepage: www.elsevier.com/locate/jocn

Tools and techniques Preoperative lumbar drainage placement for surgical cranioplasty ⇑ Henrik Giese , Jennifer Meyer, Andreas Unterberg, Christopher Beynon

Department of Neurosurgery, University of Heidelberg, Im Neuenheimer Feld 400, 69120 Heidelberg, Germany article info abstract

Article history: Objective: Intraoperative reduction of cerebrospinal fluid may be required in patients undergoing cranio- Received 19 December 2017 plasty (CP) surgery, especially in the presence of bulging cranial defects. Direct cannulation of the frontal Accepted 12 March 2018 horn of the lateral ventricle is associated with risks such as intracerebral haemorrhage or postoperative Available online xxxx leakage of cerebrospinal fluid. Here we report our initial experiences with preoperative lumbar drain (LD) placement for cranioplasty surgery in patients with bulging cranial defects. Keywords: Method: The medical records of patients who were treated with LD prior to CP surgery at our institution Cranioplasty were retrospectively analysed. Pre-, intra- and postoperative modalities, complications and outcome Decompressive craniectomy parameters are described. Lumbar drain Cerebrospinal fluid drainage Results: A total of 14 patients (mean age 46 years) were included in this analysis. The majority of patients had received decompressive craniectomy due to space-occupying cerebral infarction (64.3%) and trau- matic brain injury (14.3%). CP was performed unilaterally with autologous bone graft in 93% of cases. No complications occurred during lumbar drainage placement and the grafts were implanted without the need of dural opening and ventricle puncture. LDs were maintained for an average of 17 ± 15 h (range, 1–48 h). There were no LD related complications in the further course. Conclusion: Our initial experiences demonstrate that preoperative LD placement facilitates CP in patients with bulging cranial defects requiring surgical cranioplasty. Intraoperative puncture of ventricles for reducing cerebrospinal fluid is avoided and therefore, the risk of intracerebral haemorrhage and brain damage is reduced. Further studies are needed to evaluate the advantages of this technique as an alter- native to conventional methods of intraoperative CSF reduction. Ó 2018 Elsevier Ltd. All rights reserved.

1. Introduction However, bulging defects may also occur without edema or HC. In these cases, the CP procedure can be difficult because the cran- Decompressive craniectomy (DC) is an effective option to ioplasty cannot be adapted into the desired position. Temporary decrease life-threatening intracranial hypertension in traumatic CSF reduction may be necessary and this can be achieved through brain injury (TBI) and patients with haemorrhagic or ischemic various methods. Usually, the frontal horn of the lateral ventricle is stroke [1–4]. CP surgery with reinsertion of the autologous bone cannulated intraoperatively through insertion of an external ven- flap or implantation of alloplastic grafts is usually required in the tricular drainage (EVD) or a Cushing cannula. However, this proce- further course of respective patients. The surgical procedure is dure is associated with the risk of misplacement or intracerebral associated with high rates of complications such as wound healing haemorrhage (ICH) with subsequent postoperative neurological disorders, infections or osteolysis of the autologous bone [5,6]. This deterioration [7–10]. EVD-associated haemorrhagic complications especially applies to patients with bulging cranial defects, as punc- occur in 7.0% of cases (95% confidence interval: 4.5%–9.4%) [7,8]. ture of the ventricles to decrease the volume of intracranial cere- The use of preoperative lumbar drainage (LD) has been reported brospinal fluid for graft implantation may be necessary. for the prevention of spinal cord ischemia in patients undergoing Herniation of the cranial defect can have various reasons. thoracic endovascular aortic repair [11]. Furthermore, LD’s are fre- Cerebral edema may be present in the acute phase after TBI or quently used in scull base surgery to prevent or treat postoperative stroke. Hydrocephalus (HC) may also lead to a bulging cranial CSF fistulas, especially for extended endoscopic endonasal defect which may require implantation of a CSF shunt system. approaches with large dura defects [12,13]. The technique of LD placement is similar to that of epidural anaesthesia which is rou- tinely used, e.g. for caesarean section [14]. Therefore, the place- ⇑ Corresponding author. ment of the LD can also be performed by anaesthesiologists. To E-mail address: [email protected] (H. Giese). the best of our knowledge, its use for reduction of CSF prior to https://doi.org/10.1016/j.jocn.2018.03.008 0967-5868/Ó 2018 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Giese H et al. Preoperative lumbar drainage placement for surgical cranioplasty. J Clin Neurosci (2018), https://doi.org/ 10.1016/j.jocn.2018.03.008 2 H. Giese et al. / Journal of Clinical Neuroscience xxx (2018) xxx–xxx surgical CP has not yet been reported in the literature. The risk of During the study period, the use of LD placements was not sub- peri-procedural intracerebral haemorrhage can be minimized and ject to a specific decision tree or paradigm but was left to the dis- the craniotomy defect can be brought into the desired position cretion of the attending neurosurgeon. In general, patients were by varying the CSF flow rate (pre- and intraoperatively). considered for this procedure when a bulging cranial defect was present without signs of cerebral edema or HC (Fig. 1). Patients 2. Methods who experienced only intraoperative bulging defect, received a conventional cannulation of the frontal horn of the lateral ventricle In this retrospective study, the data of consecutive patients through insertion of an EVD or a Cushing cannula. treated with preoperative lumbar drainage prior to CP surgery at All lumbar drains (Epidural Minipack System 4, Protex/Smiths our institution were analysed. Medical, Grasbrunn, Germany) were placed by a neurosurgeon. Patient demographic information, pathology, specific risk fac- Drainage of spinal fluid was typically maintained at 10 cc per hour tors and details of the surgery and hospital course were analysed. (volume led) depending on the shape of the cranial defect (Fig. 2). Further analysis focused on time of CP and LD placement, duration The drain was removed immediately after the operation. of LD, peri- and postoperative complications and neurological out- The data were retrospectively collected from the clinical data- come (assessed by modified Ranking Scale). base and inserted into an Excel analysis. For the descriptive

Fig. 1. A+B:CT-scan of a female patient five month after an MCA-stroke and right-sided decompressive craniectomy. The cranial defect is raised over the bone margins and CP without drainage of CSF is not possible. C: CT-scan of the same patient one day after successful CP using the LD for CSF reduction.

Fig. 2. Example of an LD placement. A patient with bulging cranial defect received an LD. Due to the continuous CSF drainage the bulging defect decreased to a normal head shape and CP can be performed.

Please cite this article in press as: Giese H et al. Preoperative lumbar drainage placement for surgical cranioplasty. J Clin Neurosci (2018), https://doi.org/ 10.1016/j.jocn.2018.03.008 H. Giese et al. / Journal of Clinical Neuroscience xxx (2018) xxx–xxx 3 analysis, absolute and relative frequencies are given as mean and Table 3 standard deviation. Peri- and postoperative details of CP and LD placement. Cranioplasty Time between CE and CP days 3. Results Mean (Range) 163 (55–478) CP material n A total of 14 patients were included in this analysis. Indications Autologous 13 for DC are presented in Table 1. Patients had a mean age of 46 years PEEK 1 with a balanced gender distribution. The main indications for per- Installation CSF drainage n Pre-operative 10 forming CP after craniectomy were space-occupying cerebral Intra-operative 4 infarctions (64.3%) and traumatic brain injuries (14.3%). In the Drainage time hours majority of patients (93%), a one-sided decompression had been Mean 17 ± 15 performed and in one patient a bifrontal craniectomy had been Min 1 carried out. Patient specific risk factors are presented in Table 2. Max 48 Intra-operative details n The most common risk factors included arterial hypertension Dural injury 4 (71.4%) and nicotine abuse (28.6%). Fig. 3 shows the neurological Antibiotics n status (modified Ranking Scale) before CP surgery. Intra-operative 14 All peri- and postoperative details of CP and LD placement are Post-operative 9 Duration 3 ± 1.66 days described in Table 3. CP surgery was performed after an average Length of hospital stay days of 163 days following decompressive craniectomy. 13 patients Mean (Range) 9 (3–24)

Table 1 (93%) received an autologous implant and one patient received Characteristics of patients undergoing CP-surgery. an alloplastic implant (PEEK, Synthes, Umkirch, Germany). LD were placed prior to CP (one day before or a few hours prior to surgery) Population 14 in 8 patients and in 4 cases immediately prior surgery during gen- Gender n (%) eral anaesthesia. LDs were maintained for an average of 17 ± 15 h Male 7 (50) (range, 1–48 h). All patients received prophylactic single shot Female 7 (50) Ò Age antibiotic therapy with cefazolin (Cephazolin Fresenius Kabi ) Mean (Range) 46 y (13–69) prior to surgery. Nine patients (66,6%) received antibiotic therapy Pathology n (%) for a mean duration of 3 ± 1.66 days. The mean length of hospital Ischemic Stroke 9 (64.3) stay was nine days (range 3–24 days). Trauma 2 (14.3) Intracerebral haemorrhage 1 (7.1) In all patients, graft implantation was possible without the need Subarachnoid haemorrhage 1 (7.1) of intraoperative puncture of the ventricles. Meningitis 1 (7.1) There were no LD related complications. Four patients had signs Craniectomy n (%) of hydrocephalus in the further course following CP and were trea- One side 13 (93) Right 8 ted with implantation of a CSF shunt system. One patient devel- Left 5 oped symptomatic seizures and was subsequently treated with Bifrontal 1 (7) anticonvulsive drugs. One patient showed a bone resorption of the autologous implant after 6 month and received an alloplastic CP. The modified Ranking Scale was unchanged after surgery.

Table 2 Patient specific risk factors. 4. Discussion n (%) Smokers 4 (28.6) With the use of LD, CP was feasible without peri-procedural Arterial hypertension 10 (71.4) complications in all 14 patients with preoperatively elevated cra- Adiposity 2 (14.3) nial defects. Thus, the risk of intracerebral haemorrhage and cra- Diabetes mellitus 0 (0) Multiresistant bacteria 2 (14.3) nial CSF fistula through avoidance of intraoperative ventricular Anticoagulation CSF reduction was, at least theoretically, decreased. -Phenprocoumon 1 (7.1) A recent meta-analysis by Bauer et al. (2011) analysed a total of -Heparin 4 (28.6) 16 studies with 2428 ventriculostomy procedures [7]. ICH was -Platelet aggregation inhibitors 6 (42.8) found in 203 procedures, and 52 of these haemorrhages were

Fig. 3. Modified Ranking Scale (mRS) before cranioplastic surgery.

Please cite this article in press as: Giese H et al. Preoperative lumbar drainage placement for surgical cranioplasty. J Clin Neurosci (2018), https://doi.org/ 10.1016/j.jocn.2018.03.008 4 H. Giese et al. / Journal of Clinical Neuroscience xxx (2018) xxx–xxx deemed significant by the authors. This resulted in a cumulative However, further studies are necessary to characterise the role of rate of haemorrhage after ventriculostomy of 7.0% and a rate of sig- LD placement in CP surgery. nificant haemorrhage of 0.8%. In contrast, using LD for CSF reduc- tion, the risk of a ventriculostomy associated ICH, neurological Conflicts of interest and source of funding deterioration and other often occurring complications, such as sei- zures, can be avoided. The authors declare that they have no competing interests and In addition to the risk of haemorrhage, the correct cannulation no source of funding. of the ventricle can be difficult, especially in patients after decom- pressive craniectomy. Several studies estimated the proportion of References inaccuracy in the standard freehand technique for patients without DC to be ranged between 12.5 and 40% [15–17]. In patients treated [1] Jüttler E, Unterberg A, Woitzik J, Bösel J, Amiri H, Sakowitz OW, et al. with DC, the risk of incorrect EVD placement with subsequent Hemicraniectomy in older patients with extensive middle-cerebral-artery stroke. N Engl J Med 2014;370:1091–100. https://doi.org/10.1056/ intracerebral haemorrhage (ICH) and/or postoperative neurological NEJMoa1311367. deterioration seems even higher due to inaccurate anatomical [2] Hutchinson PJ, Kolias AG, Timofeev IS, Corteen EA, Czosnyka M, Timothy J, et al. landmarks and soft tissue displacement. Trial of decompressive craniectomy for traumatic intracranial hypertension. N Engl J Med 2016;375:1119–30. https://doi.org/10.1056/NEJMoa1605215. The main complications of LD placement include haemorrhage, [3] Cooper DJ, Rosenfeld JV, Murray L, Arabi YM, Davies AR, D’Urso P, et al. infection, material fracture and CSF fistula [11,18,19]. Esterea Decompressive craniectomy in diffuse traumatic brain injury. N Engl J Med et al.’s study reports on the usage of lumbar CSF drains during tho- 2011;364:1493–502. https://doi.org/10.1056/NEJMoa1102077. [4] Vahedi K, Vicaut E, Mateo J, Kurtz A, Orabi M, Guichard J-P, et al. Sequential- racic aortic repair [19]. They placed 1107 catheters between design, multicenter, randomized, controlled trial of early decompressive September 1992 and August 2007. The overall complication rate craniectomy in malignant middle cerebral artery infarction (DECIMAL Trial). was 1.5% (17 of 1107). No spinal haematomas were observed. Stroke; J Cereb Circul 2007;38:2506–17. https://doi.org/10.1161/ STROKEAHA.107.485235. Intracranial haemorrhage related to CSF drainage occurred in [5] Morton RP, Abecassis IJ, Hanson JF, Barber JK, Chen M, Kelly CM, et al. Timing of 0.45% (5 of 1107) of cases. The majority of haemorrhagic events cranioplasty: a 10.75-year single-center analysis of 754 patients. J Neurosurg were cerebellar (80%, 4 of 5), with one subdural haemorrhage. 2017:1–5. https://doi.org/10.3171/2016.11.JNS161917. Besides haemorrhagic complications the authors reported that 7 [6] Wachter D, Reineke K, Behm T, Rohde V. Cranioplasty after decompressive hemicraniectomy: underestimated surgery-associated complications? Clin patients developed CSF leakage. Neurol Neurosurg 2013;115:1293–7. https://doi.org/10.1016/ Due to the small sample size, our report does not allow to draw j.clineuro.2012.12.002. firm conclusions regarding the risks of complication in this patient [7] Bauer DF, Razdan SN, Bartolucci AA, Markert JM. Meta-analysis of hemorrhagic complications from ventriculostomy placement by neurosurgeons. collective. Furthermore, most complications have been reported Neurosurgery 2011;69:255–60. https://doi.org/10.1227/ for cases of longer duration of LD and higher drainage rates. Using NEU.0b013e31821a45ba. LD for CP surgery, the volume of the drained CSF can be addition- [8] Binz DD, Toussaint LG, Friedman JA. Hemorrhagic complications of ventriculostomy placement: a meta-analysis. Neurocrit Care 2009;10:253–6. ally controlled by shape and the elevation of the craniotomy defect. https://doi.org/10.1007/s12028-009-9193-0. The amount of the drainage volume should be adjusted in a way [9] Dey M, Jaffe J, Stadnik A, Awad IA. External ventricular drainage for that the defect easily falls below the bone level. Furthermore, the intraventricular hemorrhage. Curr Neurol Neurosci Rep 2012;12:24–33. https://doi.org/10.1007/s11910-011-0231-x. drainage should be removed immediately after surgery in order [10] Dey M, Stadnik A, Riad F, Zhang L, McBee N, Kase C, et al. Bleeding and to avoid the risk of epidural haemorrhage. infection with external ventricular drainage. Neurosurgery 2015;76:291–301. Four patients developed hydrocephalus after CP. This should https://doi.org/10.1227/NEU.0000000000000624. [11] Hanna JM, Andersen ND, Aziz H, Shah AA, McCann RL, Hughes GC. Results with not be regarded as a complication of the LD, but rather as a latent selective preoperative lumbar drain placement for thoracic endovascular hydrocephalus which decompensated after graft implantation. aortic repair. Ann Thoracic Surg 2013;95:1968–75. https://doi.org/10.1016/j. Nevertheless, a recently published review by Mustroph et al. athoracsur.2013.03.016. showed that the presence of a ventricular peritoneal shunt during [12] Gardner PA, Kassam AB, Thomas A, Snyderman CH, Carrau RL, Mintz AH, et al. Endoscopic endonasal resection of anterior cranial base meningiomas. CP is associated with a higher rate of overall complications (18.5%), Neurosurgery 2008;63:36–54. https://doi.org/10.1227/01. including infection and bone resorption [20]. Patients with absence NEU.0000335069.30319.1E. of a ventricular peritoneal shunt had an overall complication rate [13] Borg A, Kirkman MA, Choi D. Endoscopic endonasal anterior skull base surgery: a systematic review of complications during the past 65 years. World of 5.1%. The authors therefore recommend staged procedures Neurosurg 2016;95:383–91. https://doi.org/10.1016/j.wneu.2015.12.105. because of an increased risk of complications during simultaneous [14] Afolabi BB, Lesi FE. Regional versus general anaesthesia for caesarean section. shunt and CP surgery (complication rate simultaneous vs. staged In: Afolabi BB, editor. Cochrane Database of Systematic Reviews, vol. 10, Chichester, UK: John Wiley & Sons, Ltd; 2012, p. CD004350. doi:10.1002/ surgery, 45.7% vs. 11.6%). In our view, this also speaks in favour 14651858.CD004350.pub3. of using a LD. The LD can help to avoid a simultaneous surgery [15] Huyette DR, Turnbow BJ, Kaufman C, Vaslow DF, Whiting BB, Oh MY. Accuracy for patients with hydrocephalus, reduce the risk of complications of the freehand pass technique for ventriculostomy catheter placement: retrospective assessment using computed tomography scans. J Neurosurg and allow for a staged procedure. 2008;108:88–91. https://doi.org/10.3171/JNS/2008/108/01/0088. Further studies are clearly necessary to evaluate the potential [16] Hsieh C-T, Chen G-J, Ma H-I, Chang C-F, Cheng C-M, Su Y-H, et al. The and limitations of this technique prior to CP surgery. Nevertheless, misplacement of external ventricular drain by freehand method in emergent neurosurgery. Acta Neurol Belgica 2011;111:22–8. it represents an effective aid to facilitate CP surgery in patients [17] Toma AK, Camp S, Watkins LD, Grieve J, Kitchen ND. External ventricular drain with bulging cranial deficits, avoiding the necessity to puncture insertion accuracy: is there a need for change in practice? Neurosurgery the ventricles in order to reduce intracranial CSF. 2009;65:1197-200-1. doi:10.1227/01.NEU.0000356973.39913.0B. [18] Simmerman SR, Fahy BG. Retained fragment of a lumbar subarachnoid drain. J Neurosurg Anesthesiol 1997;9:159–61. 5. Conclusion [19] Estrera AL, Sheinbaum R, Miller CC, Azizzadeh A, Walkes J-C, Lee T-Y, et al. Cerebrospinal fluid drainage during thoracic aortic repair: safety and current management. Ann Thoracic Surg 2009;88:9–15. https://doi.org/10.1016/j. Preoperative lumbar drain placement for CSF reduction repre- athoracsur.2009.03.039. sents a potential aid in patients undergoing surgical CP. In contrast [20] Mustroph CM, Malcolm JG, Rindler RS, Chu JK, Grossberg JA, Pradilla G, et al. to classic CSF reduction through puncture of the ventricles (EVD, Cranioplasty infection and resorption are associated with presence of a ventriculoperitoneal shunt: a systematic review and meta-analysis. World Cushing cannula), the risk of brain injury and cranial CSF leakage Neurosurg 2017. https://doi.org/10.1016/j.wneu.2017.04.066. is avoided. Our initial experiences demonstrate the potential of this technique in patients with bulging cranial defects undergoing CP.

Please cite this article in press as: Giese H et al. Preoperative lumbar drainage placement for surgical cranioplasty. J Clin Neurosci (2018), https://doi.org/ 10.1016/j.jocn.2018.03.008