Online ISSN : 1687-8329


Quick Search 

July2010 Vol.47 Issue:      3 (Supp.) Table of Contents
Full Text

Role of the External Lumbar Drain in Management of CSF Leak during or after Transsphenoidal Surgery

Amr Mansy, Ahmed Kersh, Ehab Eissa


Department of Neurosurgery, Cairo University; Egypt



Background: CSF leak during or after transsphenoidal surgery is the most common annoying complication that may lead to fatal meningitis. Objective: To define the effectiveness of the external CSF lumbar drain to stop the leak either inserted preoperative, immediate postoperative or delayed postoperative with the same figure of packing to the sella turcica and sphenoid sinus. Methods: We retrospectively reviewed a series of 200 consecutive transsphenoidal surgery for pituitary adenoma, half of cases with intra-operative leak managed with packing only, the another half is managed by pack with CSF lumbar drain. Results: Leak stopped in all cases with lumbar drain except one case 19/20 (99.5%), leak stopped in this case after repacking with lumbar drain, while in cases managed with packing only, leak stopped in 13/20 (65%), leak stopped in 6/7 (85.7%) after lumbar drain, while in the seventh case, the leak did not stop after repacking with lumbar drain but stopped after lumboperitoneal shunt. Conclusion: CSF external lumbar drain markedly reduces the incidence of postoperative leak and also solves the problem of persistent postoperative leak either alone or with repacking. [Egypt J Neurol Psychiat Neurosurg. 2010; 47(3): 483-488]


Key Words: CSF rhinorrhea, pituitary adenoma, transsphenoidal surgery.


Correspondence to Ahmed Kersh, Department of Neurosurgery, Cairo University, Egypt.

Tel.:+20161036336. Email:



               Pituitary adenoma is commonly approached through transsphenoidal surgery with less morbidity especially in experienced hands1.

               One of the most annoying complication is the intraoperative CSF leak, we may not able to avoid the intraoperative leak but we try to avoid its persistence postoperatively to prevent its complications like meningitis and pneumocephalus2.

               The incidence of postoperative CSF leak after transsphenoidal surgery is 0.5-15% in reported series1,3,4. External CSF lumbar drain is a debatable tool, which can be used with packing to reduce the incidence of post-transsphenoidal leak, its usage depends on the personal preference of the surgeon, not emphasized as a rule3,5.

               In this study we aimed at comparing the effectiveness of the packing alone to the packing with lumbar drain, also we showed the effectiveness of the lumbar drain in the first group if they leaked postoperatively.





               We retrospectively reviewed a series of 200 consecutive transsphenoidal operations, all have pituitary adenoma approached via microscopic sublabial transseptal transsphenoidally all of them had done by the same surgeon in Cairo University (A.M) from January 2001 to January 2010. They were 118 women and 82 men, the age at time of TSS ranged from 25-70 years old (mean 47.5 years).

               Pathology of all lesions revealed pituitary adenoma, 22 were microadenoma (all were functioning), 178 were macroadenoma (38 were functioning, while 140 were nonfunctioning), 20 were recurrent adenoma with history of previous TSS, four of them had done  radiotherapy

               In cases without intraoperative leak, we packed the sella with gelatinous sponge and reconstructed the sellar floor with bone fragments harvested during the opening to prevent arachnoidocele.


In this study from 2001 to 2006, 20 cases (10%) with intraoperative leak were packed as follow:

§   1st layer: On the inferior surface of the arachnoid autologous fascia lata graft about 1 cm x 1 cm soaked with fibrin glue.

§   2nd layer: Is gelatinous sponge.

§   3rd layer: Is fibrin glue.

§   4th layer: Bone fragment designed to the opening of sellar floor, placed toward the sella not toward the sphenoid sinus to be rested on the edges of the floor opening.

§   5th layer: Piece of fat rolled in oxidized cellulose and soaked in fibrin glue inside the sphenoid sinus as shown in Figure (1).

               Since 2006, 20 cases with leak (10%) were packed as mentioned previously with immediate insertion of external lumbar drain  size 16 gauge to drain 60 cm CSF daily for 5 days6.

               All patients with intraoperative leak were observed postoperatively for leak after removal of nasal pack for 3 months to detect the delayed leak.

Figure (2) shows an allogram of the management of the leak during the transsphenoidal surgery.

Data were expressed as number and percentage and tabulated using Microsoft EXCEL® 2007.








Figure 1. The following pictures showed three layers during reconstruction of the sella and the sphenoid sinus after transsphenoidal surgery: Slide (A) show the reconstruction of the sellar floor by the harvested bon, slide (B) show the gelatinous sponge layer inside the sella, slide (C) show piece of fat rolled  in oxidize cellulose inside the sphenoid sinus.





Intraoperative leak was encountered in 40/200 cases (20%), in all cases the leak is detected immediately, with higher incidence in macroadenomas than microadenomas, especially those with suprasellar extension (22%), as we were more aggressive in macroadenomas to remove the mass totally, the most trigger cause was invasion or adherence to diaphragm sella ,it was mostly teared when we tried to remove the last layer adherent to the thinned out diaphragm with postoperative radiology to these cases showed no residual mass, so adhesion to the diaphragm with the aim of total excision, mostly ended by arachnoidal tear.

               Incidence is higher in non-functioning adenomas (21.4%) than the functioning ones (16.7%), higher incidence is related to the size and suprasellar extension7 as shown in Table (1).

               Incidence also is apparently higher in recurrent adenomas (35%) mostly due to distorted anatomy and adhesions, it is higher in recurrent cases with radiotherapy (50%) due to decreased vascularity that impend the healing process. Recurrent tumors, radiotherapy and aggressive resection of the tumor, aiming total excision were the most common risk factors leading to postoperative rhinorrhea in this study. The postoperative leak were detected at different times, in the only case with leak after lumbar drain, leak was detected after 2 weeks, in cases managed firstly without drain, 3/7  leak was detected after one week, 4/7 leak was detected after pack removal. In all cases with postoperative leak we did CT brain before any management to exclude hydrocephalus or pneumocephalus, hydrocephalus was not present, while pneumocephalus was present in 3 cases, with larger amount in the case in which we used the lumbar drain, but with meticulous repacking and lumbar drain, the air disappeared gradually as shown in (Figure 3).


Table 1. Showed the relation of the size, functioning or non-functioning adenoma to number of cases with leak or without leak during transsphenoidal surgery.



Without leak

With leak





Macroadenoma (178)

With SSE 100





Without SSE 78










Cushing (15)





Prolactinoma (10)





Acromegalic (35)





Non-functioning (140)





Recurrent (20)





Radiotherapy (4)









Figure 2. Management of the leak during the transsphenoidal surgery.





Figure 3. Follow up CT brain for pituitary adenoma operated upon transsphenoidally with intraoperative leak managed by packing with CSF external lumbar drain, which led to large amount of pneumocephalus due to non meticulous repair, pneumocephalus markedly improved with drain after meticulous repair (A) show axial CT brain with large amount of pneumocephalus immediately after starting leak, (B) axial CT brain of the same patient after meticulous repacking with lumbar drain, complete recovery of the pneumocephalus.





               Postoperative CSF leak is a common complication when the pituitary adenoma is approached transsphenoidally and encountered in reported series 0.5-15%5,8, in this series the incidence was 4%.

               The risk factors that increase the incidence of the intraoperative leak includes: large adenomas especially those with suprasellar extension ,recurrent adenomas, previous radiotherapy, preoperative leak and aggressive surgical maneuver either in non experienced hands or in experienced hands aiming the total excision especially when trying to remove the adherent layer of the tumor to the diaphragma sella ,the main risk factor for the postoperative leak is the intraoperative leak, so the postoperative leak incidence increase by the same risk factors of the intraoperative leak, non meticulous repair of the sella is another important factor8,9.

               Many ways have mentioned in repair if intraoperative CSF leak occurred such as placement of nonabsorbable or slowly absorbable substance into the sella/fat, muscle, bioglue collagen sponge, fibrin glue and sellar floor reconstruction with autologous, heterologous or synthetic material: cartilage, bone, fascia lata, titanium mesh1,9,10.

               In this series we used fascia lata, gelatinous sponge, fibrin glue, fat with oxidize cellulose.

               Fat was used to pack the sphenoid sinus not inside the sella to avoid the interpretation of the sellar content on the postoperative MRI, as the reabsorption takes longer time in comparison to bioabsorpable packing such as gelatinous sponge, also chiasmal compression is a rare complication that may occur when we pack the sella with fat11.

               Fibrin glue produce a thin rim of low density which less confounding in postoperative MRI interpretation than fat graft, but it is a costy material12.

               Performance of the valsalva's maneuver at the end of the surgery is mandatory done either in cases without apparent CSF leak to be sure that no leak or in cases with intraoperative leak to document the good sealing and effectiveness of the sellar reconstruction13.

               The use of external lumbar drain is in debate, in reported series it may be used preoperatively in selected cases when they expect an intraoperative leak due to large adenomas and or when they intend to use saline infusion to facilitate the descent of the suprasellar part of the adenoma ,but this increase the risk of the intraoperative leak(11) ,  some authors state that the drain may hide the earlier detection of the CSF leak5. Others recommended its use if intraoperative leak occurred, in this work, the lumbar drain had a big role in reducing the leak from (35%) to (,5%), it also had a big role in the management of the postoperative leak in the first group managed with packing only, either used alone or with repacking.

               The effectiveness of lumbar drain in this work is 99.5% which reached 100% after repacking and lumbar drain to the lonely case leaked after transsphenoidal surgery, this result is near or equal to reported efficacy of Shapiro and Scully13,14.

No complications developed from the lumbar drain such as nerve root irritation, meningitis, CSF overdrainage, but its use increase the time of stay in the hospital and leads to tension pneumocephalus only in patients with non meticulous packing13,15.

In spite that it is effective with repacking in management of tension pneumocephalus developed with delayed CSF leak.

               All patients were covered by prophylactic antibiotics, the whole time of drain or pack.

               In our experience combined duroplasty with fascia lata graft, fibrin glue, gelatinous sponge with fat inside the sphenoid sinus only followed by external lumbar drain, to drain CSF 60 cm/day for 5 days is highly efficient management to prevent persistent CSF leak after transsphenoid surgery.


[Disclosure: Authors report no conflict of interest]




1.      Ciric I, Ragin A, Craig B. Complications of transsphenoidal surgery: results of national survey, review of the literature, and personal experience. Neurosurgery. 1997; 40: 225-37.

2.      Sawka AM, Aniszewski JP, Young Jr WF, Nippoldt TB, Yanez P, Ebersold MJ. Tension pneumocranium, a rare complication of transsphenoidal surgery: Mayo Clinic experience 1976-1998. J Clin Endocrinol Metab. 1999; 84: 4731-4.

3.      Black PMCL, Zervas NT, Candia GL. Incidence and management of complications of transsphenoidal operation for pituitary adenomas. Neurosurgery. 1987; 20: 920-4.

4.      Shiley SG, Limonadi F, Delashaw JB, Barnwell SL, Andersen PE, Hwang PH, et al. Incidence, etiology and management of cerebrospinal fluid leak following transsphenoidal surgery. Laryngoscope. 2003; 113: 1283-8.

5.      Jane JA Jr, Thapar K, Kaptain GJ, Maartens N, Lows ER Jr. Pituitary surgery: transsphenoidal approach. Neurosurgery. 2002; 51: 435-44.

6.      Alejeindra T, Rabadan C, Ruggeri S. Pituitary tumor: our experience in the prevention of post-operative cerebrospinal fluid leaks after transsphenoidal Surgery. 2009; 93: 127-31.

7.      Nishioka H, Haraoka J, Ideda Y. Risk factors of cerebrospinal fluid rhinorrhea following transsphenoidal surgery. 2005; 147: 1163-6.

8.      Laws ER Jr, Fode NC, Redmond MJ. Transsphenoidal surgery following unsuccessful prior therapy. J Neurosurg. 1985; 63: 823-9.

9.      Seiler RW, Mariani L. Sellar reconstruction with resorbable vicryl patches, gelatin foam, and fibrin glue in trans-sphenoidal surgery: a 10 years experience with 376 patients. J Neurosurgery. 2000; 93: 762-5.

10.    Kaptain GA, Vincent DA, Laws ER Jr. Cranial base reconstruction after transsphenoidal surgery with bioabsorbible implants: technical notes. Neurosurgery. 2001; 48: 232-4.

11.    Sade B, Mohr G, Frenkic LS. Management of intraoperative cerebrospinal fluid leak in transnasal transsphenoidal pituitary microsurgery: use of postoperative lumbar drain  and sellar reconstruction without fat packing. Acta Neurochir (wien). 2006; 148: 13-19.

12.    Steiner E, Knosp E, Herold CJ, Kramer J, StigIbaver R, Staniszewski K, et al. Pituitary adenomas: findings of postoperative MRI finding. Radiology. 1992; 85: 521-7.

13.    Shapiro SA, Scully T. Closed continuous drainage of cerebrospinal fluid via a lumbar subarachnoid catheter for treatment or prevention of cranial/spinal cerebrospinal fluid fistula. Neurosurgery. 1992; 30: 241-5.

14.    Esposito F, Dusick JR, Fatemi N, Kelly DF. Graded repair of cranial base defects and cerebrospinal fluid leaks in transsphenoidal surgery. Neurosurgery. 2007; 60: 295-304.

15.    Acikbas SC, Akyuz M, Kazan S, Tuncer R. Complications of closed continuous lumbar drainage of cerebrospinal fluid. Acta Neurochir (Wien). 2002; 144: 475-80.



الملخص العربى


دور الدرنقة القطنية الخارجية لسحب السائل النخاعي في علاج تسرب السائل النخاعي

أثناء أو بعد جراحات الغدة النخامية من الأنف


تسرب السائل النخاعي من الأنف بعد جراحات الغدة النخامية هي أشهر المضاعفات المقلقة بعد هذه الجراحة. دور هذا البحث هو إثبات فعالية الدرنقة الخارجية لسحب السائل النخاعي من الظهر، سواء تم تركيبها قبل الجراحة مباشرة أو بعد الجراحة أو متأخرة عند الحدوث المتأخر للسائل النخاعي في علاج هذا التسرب النخاعي مع توحيد طريقة حشو تجويف الغدة النخامية والجيب الهوائي العنكبوتي. وقد تم نظر مائتي حالة من جراحات الغدة النخامية من الأنف، كل الحالات التي تسرب فيها السائل النخاعي أثناء الجراحة تم علاجهم بالحشو بنفس الطريقة مع إضافة درنقة قطنية خارجية لسحب السائل النخاعي في النصف الأخير من الحالات وكانت نسبة منع التسرب بعد الجراحة في حالات الدرنقة 99.5% مقابلة لـ65% في النصف التي تم عمل حشو فقط دون تركيب درنقة.

وقد خلصت هذا الدراسة إلى أن الدرنقة القطنية الخارجية لسحب السائل النخاعي تقلل بنسبة كبيرة حدوث تسرب هذا السائل النخاعي بعد جراحات الغدة النخامية من الأنف وهي أيضا ذات دور أساسي في علاج التسرب للسائل النخاعي في الحالات التي لم يتم فيها تركيب الدرنقة بعد الجراحة مباشرة.

2008 � Copyright The Egyptian Journal of Neurology,
Psychiatry and Neurosurgery. All rights reserved.

Powered By DOT IT