RESEARCH—HUMAN—CLINICAL STUDIES

Role of Craniotomy Repair in Reducing

Postoperative Headaches After a
Retrosigmoid Approach
Mario K. Teo, MBBS, MRCS
Department of Neurosurgery,
The University of Dundee,
Ninewells Hospital and Medical School,
Dundee, Scotland
M. Sam Eljamel, MD, FRCS
(Ed, Ir, SN)
Department of Neurosurgery,
The University of Dundee,
Ninewells Hospital and Medical School,
Dundee, Scotland
Reprint requests:
M. Sam Eljamel, MD, FRCS(Ed, Ir, SN),
Consultant Neurosurgeon,
Department of Neurosurgery,
Ninewells Hospital & Medical School,
Dundee DD1 9SY, Scotland.
E-mail: m.s.eljamel@dundee.ac.uk
Received, September 27, 2009.
Accepted, April 19, 2010.
Copyright ª 2010 by the
Congress of Neurological Surgeons
BACKGROUND: The retrosigmoid (RS) approach provides an excellent access corridor to
the cerebellopontine angle. However, 80% of patients experience headaches after RS
approaches.
OBJECTIVE: We reviewed our prospective database to determine the risk factors
influencing headaches after RS procedures.
METHODS: From 2003, craniotomy, instead of craniectomy, became our standard
approach for RS procedures. Patients’ demographic, management, and outcome data
were collected prospectively. We also retrospectively analyzed similar data collected
between 2000 and 2003 to compare headache outcomes after RS approaches. Subgroup
analysis of data was performed to identify other risk factors contributing to post-
operative headaches.
RESULTS: Of 105 patients (mean age, 56 years; 43 men; 62 women) who underwent RS
surgery, 30 underwent craniectomy and 75 underwent craniotomy. There were 57
vestibular schwannomas, 40 microvascular decompressions, and 8 other procedures. The
patients’ age, sex, pathological diagnosis, and length of hospital stay were not statis-
tically different in the 2 subgroups. At discharge, postoperative headache was observed
in 43% of patients (13/30) after craniectomy and 19% of patients (14/75) after craniotomy
(P = .01). The incidence of headache decreased with further follow-up; 10% of patients
(3/30) who underwent craniectomy and 1% of patients (1/75) who underwent crani-
otomy still had headache at 12 months of follow-up.
CONCLUSION: Patients who underwent the RS approach with craniotomy had a signifi-
cantly lower rate of headache at discharge than did those who underwent craniectomy.
These patients continued to have a lower incidence of headache in the long term.
KEY WORDS: Functional outcome, Postoperative headache, Retrosigmoid surgery

Neurosurgery 67:1286–1292, 2010 DOI: 10.1227/NEU.0b013e3181f0bbf1 www.neurosurgery-online.com

T
he retrosigmoid (RS) approach provides an excellent ex-
posure and surgical corridor to the cerebellopontine angle
(CPA) to remove vestibular schwannoma with hearing
preservation and perform microvascular decompression (MVD) of
the trigeminal nerve in patients with trigeminal neuralgia, MVD
of the facial nerve in patients with hemifacial spasm, and perform
MVD of other lower cranial nerves. The reported incidence of
headaches after the RS approach is well documented in the lit-
erature; it varies from 30 to 80%1-4 after RS excision of vestibular
schwannoma and occurs in up to 60% of patients after MVD.5
ABBREVIATIONS: CPA, cerebellopontine angle; MVD, microvascular
decompression; RS, retrosigmoid; RSC, retrosigmoid craniectomy;
RSR, retrosigmoid craniotomy with repair
The causes of these headaches have not been well characterized
and remain controversial. Some authors hypothesized that dural
adhesions between the dura and cervical suboccipital musculature
are the most likely cause.6,7 Therefore, it was proposed that
meticulous closure of the dura and cranioplasty after RS cra-
niectomy (RSC) results in fewer postoperative headaches.8,9
However, other surgical teams failed to demonstrate a beneficial
effect on RS headaches after cranioplasty.1,5
Injury to the greater and lesser occipital nerves also has been
proposed as a causative factor, as these nerves may be damaged
directly during skin incision, during retraction of the edges of the
surgical incision with self-retaining retractors, or during suturing
of the incision.10-12 Another possible cause that has been proposed
is chemical meningitis from bone dust contaminating the sub-
arachnoid space during intradural drilling of the internal auditory

1286 | VOLUME 67 | NUMBER 5 | NOVEMBER 2010 www.neurosurgery-online.com

Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.


canal.1,13 We performed this study to determine the incidence of
and risk factors for headache after RS procedures to establish how
to minimize their occurrence.
PATIENTS AND METHODS
From 2003 onward, craniotomy (with bone flap replacement), instead
of craniectomy, became our standard approach to RS procedures. This
change was based on the results of an audit we performed at the end of
2002, in which there was a trend of reduced incidence of postoperative
cerebrospinal (CSF) leak with craniotomy. Our technique involved
making a single burr hole as close as possible to the junction of the
transverse and sigmoid sinuses using image-guided surgery technology
and a craniotome. The bone dust was collected to repair the burr hole at
the end of the procedure. We aimed to make the bone flap as small as
possible for the purpose of the operation. The dura was then dissected off
the bone, and the bone flap was raised using the craniotome. The dura
was opened in a Y-shaped incision anchoring the dural flaps superiorly
and laterally. At the end of the procedure, the dura was sutured together
without artificial duraplasty. The bone flap was sutured to the dura with
a tent suture (Figure 1). The bone flap was not fixed to the surrounding
skull bone. Because the bone flap edge was bevelled during the crani-
otomy, it did not sink into the craniotomy opening; rather than suturing
it to the dura, we allowed the bone flap to rise outward with the dura.
The bone dust was used to repair the burr hole and the craniotomy bony
defect between the bone flap and the skull.
Patients’ demographics, management details, and outcome for at least
12 months were collected prospectively. Headache outcome was recorded
at discharge and at 3 months, 6 months, and 1 year after discharge. Severity
of headache was graded according to a 10-point visual analog scale. For
analysis, visual analog scale score of 3 or lower was defined as mild
headache, 4 to 6 as moderate headache, and 7 to 10 as severe headache.
Mild headache was defined as a headache with which the patient can live
with without medication; moderate headache required regular medication;
and severe headache required frequent attention from specialist pain
services and affected the patient’s quality of life.
Data were also collected for patients who underwent RS surgery
between 2000 and 2003, when RSC was performed in some patients (30
patients). We compared those who had RS craniotomy with repair (RSR)
with those who had RSC in headache outcome and possible risk factors.
All data relative to RSR were analyzed to identify other risk factors
contributing to postoperative headaches.
Only patients undergoing purely RS surgery were included; if an RS
approach was used in combination with another approach, those patients
were excluded (6 patients).
Statistical analysis with Fisher’s exact test was used to compare the
incidence of postoperative headaches in the 2 subgroups and in subgroup
analysis. A P value of less than .05 was considered statistically significant.
A logistic regression analysis model using all risk factors was used to
establish the odds ratios of each variable.
Follow-up Protocol
Patients who underwent RS excision had a magnetic resonance im-
aging (MRI) scan with gadolinium enhancement postoperatively, within
6 months of discharge, and annually thereafter. Patients who underwent
MVD did not have any routine postoperative imaging. Those who
relapsed after initial trigeminal neuralgia response to MVD (18%) had an
MRI scan (constructive interference in steady state) to reveal any
NEUROSURGERY
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
CRANIOTOMY REPAIR REDUCES HEADACHES AFTER RETROSIGMOID APPROACH
recurrence of vascular compression. Both patient groups were followed-
up at 3 months, 6 months, 12 months, and annually thereafter. Patients
who underwent MVD were administered carbamazepine or gabapentin
for 2 weeks after surgery.
RESULTS
From January 2000 to March 2008, 111 patients underwent
RS surgery. Six patients were excluded because they underwent
RS surgery in addition to another surgical approach. The mean
age of the remaining 105 patients was 56 years; 43 were men and
62 were women. Thirty patients had RSC, and 75 underwent
RSR. The RS approach was used to excise vestibular schwannoma
in 57 patients, to perform MVD in 40 patients, and to excise
other CPA lesions in 8 patients (3 CPA meningiomas, 3 CPA
arachnoid cysts, 2 CPA epidermoid cysts).
Table 1 summarizes the demographic and clinical character-
istics of the 105 patients in the study. All 57 patients with
vestibular schwannoma preferred microsurgery for a median
tumor size of 2 cm (range, 1-4 cm). None of these patients
required stereotactic radiosurgery, and 96% had House-Brack-
mann grade 1 facial function postoperatively and a 56% rate of
hearing preservation. Of the 40 patients who underwent MVD,
82% had control of their symptoms at 5 years, and no hearing loss
was reported in any of these patients. No other cranial nerve
deficits were reported in this series. MVD lasted an average of 120
minutes (range, 90-180 minutes), and vestibular schwannoma
excision lasted an average of 5 hours (range, 4-6 hours). In pa-
tients with vestibular schwannomas, 80% of cases required
drilling of the internal auditory meatus to expose the lateral limit
of the tumor. Four patients had hydrocephalus at presentation,
and 1 patient required a permanent CSF diversion procedure.
Six patients were excluded because of combined surgical ap-
proaches, which included resection of glomus jugulare tumor (3
patients), clival meningioma (2 patients), and foramen magnum
melanocytoma (1 patient).
There were no statistically significant differences between the
RSR and RSC groups in age, sex, side, and underlying diagnosis.
The size of the RS approach differed between the 2 groups;
33% of the RSRs (25/75) were less than 3 cm in diameter,
compared with 10% of RSCs (3/30) (P = .03). The remaining
bone flaps measured 3 to 6 cm in 67% of RSRs (50/75) and 90%
of RSCs (27/30) (P = .02).
Wound CSF leakage or CSF rhinorrhea was observed in 6.6%
of RSRs (5/75) and 20% of RSCs (6/30) (P = .09).
At discharge, postoperative headache was observed in 43% of
patients who underwent RSC (13/30) and 19% of those who
underwent RSR (14/75) (P = .01). The incidence of headache
decreased with follow-up, and at 12 months postoperatively, 10%
of patients in the RSR group (3/30) and 1% of patients in the
RSC group (1/75) experienced headache (P = .1) (Figure 2).
All patients in the RSC subgroup had mild headaches post-
operatively. In the RSC subgroup, of the 14 patients with
headache at discharge, 2 had moderate headache (visual analog
VOLUME 67 | NUMBER 5 | NOVEMBER 2010 | 1287
TEO AND ELJAMEL

FIGURE 1. Intraoperative photographs showing surgical technique of retrosigmoid craniotomy with repair (RSR). A, execution
of preoperative plan demonstrating the target volume head-up display on the scalp (1) and skin fiducial (2). B, location of burr
hole (3). C, measuring the size of bone flap with flexible ruler (5). D, replacing the bone flap (4) with the suture, burr hole filled
with bone dust (6), and the space created by drilling laterally (7), which will also be filled with bone dust.

scale score, 4–6) and the remaining patients had mild headache.
At 3 months and 6 months of follow-up, 1 patient continued to
have moderate headache, which eventually improved to mild
headache at 12 months of follow-up.
Long-term follow-up data (mean follow-up, 5.7 years) demon-
strated that after 1 year postoperatively, there was no change in the
incidence of headaches. However, 18% of patients who underwent
MVD (8 patients) were taking medication to control their trigeminal
neuralgia because of relapse. None of the 65 patients who had repeat
MRI scans postoperatively demonstrated MRI artifact at the bone
flap site, making image interpretation straightforward.
Subgroup Analysis
From January 2004 to March 2008, 56 patients underwent
RSR. The incidence of postoperative headache in this cohort was
13% (7/56) at discharge, 7% (4/56) at 3 months, 2% (1/56) at
6 months, and none at 1 year of follow-up.
Eighty percent of the patients (45/56) were younger than 65
years, and 61% of them (34/56) were women. Forty-eight per-
cent (27/56) had vestibular schwannomas, 38% (21/56) un-
derwent MVD for trigeminal neuralgia, and 14% (8/56) had
other conditions. As shown in Table 2, sex, age, and underlying
diagnosis did not result in a statistically significant increase in the
incidence of postoperative headache in this subgroup.
Of the 27 patients with vestibular schwannomas, 52%
(14/27) underwent intraoperative drilling of the internal
auditory meatus for the removal of intracanalicular tumor
extension. Of these patients, 14% (2/14) had postoperative
headaches compared with 15% of patients (2/13) who did
undergo intraoperative drilling. Therefore, intraoperative

1288 | VOLUME 67 | NUMBER 5 | NOVEMBER 2010 www.neurosurgery-online.com

Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.

 CRANIOTOMY REPAIR REDUCES HEADACHES AFTER RETROSIGMOID APPROACH

TABLE 1. Comparative Data Between Retrosigmoid Craniectomy TABLE 2. Subgroup Analysis of Risk Factors for Postoperative
and Retrosigmoid Craniotomya Headache After Retrosigmoid Craniotomy With Repaira
Patient Data Craniectomy Craniotomy P Value Headache at Discharge
No. operations 30 75 Yes, No. No, No.
Age, y Risk Factor Patients (%) Patients (%) P Value
Average 66.3 53 ..05
Age, y
Median (range) 66 (36-82) 53 (20-78)
,65 5 (11%) 40 (89%) 0.42
Length of hospitalization, d
.65 2 (18%) 9 (82%)
Average 9.2 8.0 ..05
Sex
Median (range) 7 (3-30) 6 (2-30)
Female 4 (12%) 30 (88%) 0.57
Male:female ratio 0.76 (13:17) 0.66 (30:45) ..05
Male 3 (14%) 19 (86%)
(no. patients)
Type of surgery
Right:left ratio 1 (15:15) 0.97 (37:38) ..05
Acoustic neuroma excision 4 (15%) 23 (85%) 0.86
(no. patients)
Microvascular decompression 2 (10%) 19 (90%)
Type of operation
Other 1 (12%) 7 (88%)
Microvascular decompression 10 (33%) 30 (40%) ..05
Intraoperative drilling
Acoustic neuroma excision 20 (67%) 37 (49%) ..05
Yes 2 (14%) 12 (86%) 0.67
Other 0 (0%) 8 (11%) ..05
No 2 (15%) 11 (85%)
Size of bone flap
Size of craniotomy
,3 cm 3 (10%) 25 (33%) .03b
,3 cm 1 (5%) 20 (95%) 0.17
3-6 cm 27 (90%) 50 (67%) .02b
3-6cm 6 (17%) 29 (83%) 0.17
.6 cm 0 (0%) 0 (0%)
Postoperative CSF leak
CSF leak 6 (20%) 5 (6%) .09
Yes 3 (75%) 1 (25%) 0.005b
Headache
No 4 (8%) 48 (92%)
Discharge 13 (43%) 14 (18.6%) .01b
Postoperative meningitis
3 mo 3 (10%) 5 (6.6%) .85
Yes 3 (100%) 0 (0%) 0.001b
6 mo 3 (10%) 2 (2.6%) .26
No 4 (8%) 49 (92%)
12 mo 3 (10%) 1 (1.3%) .12
Postoperative wound infection
a Yes 2 (100%) 0 (0%) 0.01b
CSF, cerebrospinal fluid.
b
Statistical significance. No 5 (9%) 49 (91%)
a
CSF, cerebrospinal fluid.
b
Statistical significance.
drilling was not shown to be a risk factor for postoperative
headaches in this study (P = .67).
Of patients with RSR less than 3 cm in diameter, 5% (1/21)
(6/35) with bone flap 3-6 cm in diameter. This difference was not
had postoperative headache compared with 17% of patients
statistically significant (P = .17).
The incidence of CSF leak, CSF infection, and wound
infection was 7% (4/56 patients), 5% (3/56 patients), and 3.5%
(2/56 patients), respectively.
Postoperative CSF leak, CSF infection, and wound infection
were significantly associated with postoperative headaches
(Table 2). Seventy-five percent of patients with postoperative
CSF leak (3/4) and 100% of patients with CSF infection
or wound infection had postoperative headache (P = .005,
P = .001, P = .01 respectively).
Logistic regression analysis of all potential variables was used to
predict the odds ratios of independent variables. We included
infection, CSF leak, size of craniotomy, and previous headache in
the model. The odds ratio was 6.34 for infection, 1.18 for CSF
leak, 7.5 for size of craniotomy, and 0 for previous headache.

FIGURE 2. Graph comparing postoperative headache in patients who DISCUSSION

underwent RSR vs those who had retrosigmoid craniectomy. There were 105
patients in total. The RS approach to the CPA provides an important surgical
corridor to remove lesions in the CPA. In our series, resection of

NEUROSURGERY VOLUME 67 | NUMBER 5 | NOVEMBER 2010 | 1289

Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.

TEO AND ELJAMEL
vestibular schwannoma with the intention of hearing preservation
and MVD of trigeminal neuralgia were the main indications for
this approach. With the introduction of the surgical microscope
and improvement of microsurgical techniques, the traditional
markers of surgical success (extent of tumor removal, cranial
nerve preservation, morbidity, and mortality) are no longer ad-
equate outcome parameters; more detailed quality-of-life meas-
ures need to be reported including symptoms of headaches,
dizziness, and numbness. The main disadvantage of the RS ap-
proach has been the relatively high incidence of postoperative
headache.1-5 A recent large-scale questionnaire involving 1657
patients who underwent surgery for vestibular schwannoma
surgery was coordinated by the American Acoustic Neuroma
Association, and postoperative headache was reported in one-
third of the respondents. Furthermore, those who underwent the
RS approach were significantly more likely to report their worst
postoperative headache as severe (82%) compared with the
translabyrinthine (75%) and middle fossa approaches (63%).3
This was not confirmed in our prospective study. The majority
of our patients had mild headaches after RS, and at 12 months of
follow-up, their headaches were not significant. Only 1 patient
of our 105 patients (,1%) continued to have moderate headache
requiring cranioplasty and chronic pain team input.
Surveys often suffer from low response return and bias in-
troduced by retrospective methodology. Our study was not
randomized between craniectomy and craniotomy, and this was
a limitation. Nevertheless, both procedures were used up to 2003
(30 craniectomies and 19 craniotomy), and we observed a re-
duction in the incidence of postoperative CSF leaks in the cra-
niotomy group (6/30 vs 2/20).
In the past, RSC was performed because of its technical ease
in this region compared with RSR and the assumption that re-
placing the bone flap serves only as cosmetic repair of a small
defect in a region of the cranium that is often invisible and
covered by hair. However, because of the high incidence of
headaches after RS approaches, surgical teams sought to un-
derstand the mechanisms involved. A probable mechanism for
headache after RSC is adhesions of cervical musculature to the
exposed dura at the craniectomy site. The dura of the posterior
fossa is richly innervated and capable of producing headache as
a result of traction during neck motion.7,8 Several authors have
described cranioplasty with calvarial bone graft, methyl meth-
acrylate, or titanium mesh-acrylic in an attempt to circumvent
this process,14-16 which resulted in a lower rate postoperative
headaches after cranioplasty.8 However, the benefit of cranio-
plasty was questioned in subsequent clinical studies that failed to
show any reduction in the incidence of postoperative headaches
in such patients.1,5 It also has been suggested that the cranioplasty
material may increase local tissue reaction and potentially lead to
a higher incidence of headache.5 In our series, the difference
between the 2 subgroups at discharge favored cranioplasty repair,
but the severity of headache was mild in the RSC subgroup
compared with RSR subgroup. However, persistence of headaches
at 1 year was 10 times more common after RSC compared with
1290 | VOLUME 67 | NUMBER 5 | NOVEMBER 2010
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
RSR. Our findings support the hypothesis that early after RSR,
headaches are more likely to be severe because of tissue reaction,
but once this subsides, long-term headache outcomes are better
with RS repair.
Not all neurosurgeons perform RSR. Since the senior author
(MSE) began performing craniotomy instead of craniectomy in
RS surgery, the incidence of immediate postoperative headaches
in our cohort has decreased significantly, from 43 to 19%. Within
12 months, 10% of patients who underwent RSC and 1% of
those who underwent RSR were still having headaches. In 2001,
Sepehrnia and Knopp reported no headaches in their 75 patients
who underwent osteoplastic RSR for acoustic neuroma.17
Injury to the greater and lesser occipital nerves also has been
proposed as a potential cause of headache after RS surgery. These
nerves may be injured directly during the incision or secondarily
by retraction of the incised soft tissue. In our series, with more
craniotomy flaps having a relatively smaller size compared with
craniectomy, the need to retract soft tissue was minimized. This
may also have contributed to a lower incidence of postoperative
headache in this subgroup.
Another proposed cause of postoperative headache is chemical
meningitis caused by bone dust contaminating the subarachnoid
space of the posterior fossa. Although the exact mechanism is not
well elucidated, chronic meningeal irritation from bone dust
liberated during drilling of the internal auditory canal is thought
to contribute to persistent postoperative headache in some pa-
tients.10,13 Therefore, aggressive irrigation and suctioning of
debris during intradural drilling have been proposed to remove
bone dust.1,8 Our series did not replicate findings of intra-
operative drilling increasing the incidence of headache after RS
surgery; 14% of patients who had intraoperative drilling, and
15% of those who did not, developed headache. We used con-
tinuous automated irrigation and suction during drilling and
succeeded in removing irritant bone dust in these patients.
The incidence of CSF leak was 6% in patients who underwent
RSR compared with 20% in patients who underwent RSC. The
size of the cranial opening in the RSR subgroup was less than
3 cm in diameter in 33% of patients compared with 10% of
patients in the RSC subgroup. This may have contributed to
the higher incidence of CSF leak in the RSC subgroup. However,
there was no correlation between the size of bone opening and the
incidence of CSF leak (9/77 .3 cm vs 2/28 ,3 cm, P . 0.05).
There was a correlation between the size of craniotomy and
subsequent headache (odds ratio, 7.5).
Replacing the bone flap was associated with a lower risk of CSF
leak in the postoperative period and therefore a lower risk of
wound or CSF sepsis, both of which were significant risk factors
for postoperative headache (odds ratio, 6.34). The incidence of
postoperative CSF leak (7%), CSF infection (5%), and wound
infection (3.5%) was comparable to that reported in other se-
ries.14 Although we demonstrated on a prospective basis that RSR
is better than RSC to reduce the incidence of headache after RS
surgery, and we identified CSF leakage and sepsis as important
adverse prognostic factors for postoperative headache, our study
www.neurosurgery-online.com

had several limitations: a sample size of 105 patients is relatively
small for meaningful subgroup analysis, this study was not ran-
domized between the 2 techniques, and increased experience may
have played a role in the RSR outcome. However, all procedures
were performed by the senior author, who had performed this
surgery for 5 years before the start of the study; it is unlikely that
learning curve was a significant factor in this study; and patients
in the study group had several underlying conditions, so the
approach used may have contributed to the outcome.
Patients who underwent RS excision were not taking pain
medication preoperatively, in contrast to patients who underwent
MVD for trigeminal neuralgia and those who underwent MVD
and continued their medications for 2 weeks postoperatively
before tapering them. However, patients who underwent MVD
were equally distributed between the 2 subgroups (RSC and
RSR) and therefore were unlikely to have contributed to the
difference in the incidence of headaches between the 2 study
groups. Similarly, both study groups received corticosteroids
during the perioperative period, which is unlikely to have played
a significant role in the difference. However, the use of steroids,
carbamazepine, and gabapentin perioperatively probably reduced
the overall incidence of headaches in the 2 groups. Headaches
may occur at a later date after surgery. We have not observed any
change in the long-term headache outcome in our cohort with
a mean follow-up period of 5.7 years. We also observed that by
not fixing the bone flap to the skull, we avoided metallic sus-
ceptibility artifacts on subsequent MRI scans.
CONCLUSIONS
RSR improved outcome after RS surgery by significantly re-
ducing the incidence of headache at discharge and at 1 year.
Postoperative CSF leak, CSF sepsis, size of craniotomy, and wound
infection are important risk factors for postoperative headache.
Despite popular belief that intraoperative drilling is a major risk
factor for headache after RS surgery, we did not find a correlation.
RSR is a better technique in RS surgery until a sufficiently powered
randomized controlled study demonstrates otherwise.
Disclosure
The authors have no personal financial or institutional interest in any of the
drugs, materials, or devices described in this article.
Acknowledgments
We thank Dr Aposlous Samelis, Dr A Quin, and Dr J Bowness
who assisted in data collection at the Department of Neurosurgery
while this study was under way.
REFERENCES
1. Catalano PJ, Jacobowitz O, Post KD. Prevention of headache after retrosigmoid
removal of acoustic tumors. Am J Otol. 1996;17(6):904-908.
NEUROSURGERY
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
CRANIOTOMY REPAIR REDUCES HEADACHES AFTER RETROSIGMOID APPROACH
2. Hanson MB, Glasscock ME III, Brandes JL, Jackson CG. Medical treatment of
headache after suboccipital acoustic tumour removal. Laryngoscope.
1998;108(8 Pt 1):1111-1114.
3. Ryzenman JM, Pensak ML, Tew JM Jr. Headache: a quality of life analysis in
a cohort of 1657 patients undergoing acoustic neuroma surgery, results from the
acoustic neuroma association. Laryngoscope. 2005;115(4):703-711.
4. Somers T, Offeciers FE, Schatteman I. Results of 100 vestibular schwannoma
operations. Acta Otorhinolaryngol Belg. 2003;57(2):155-166.
5. Lovely TJ, Lowry DW, Jannetta PJ. Functional outcome and the effect of cra-
nioplasty after retromastoid craniectomy for microvascular decompression. Surg
Neurol. 1999;51(2):191-197.
6. Kimmel DL. Innervation of spinal dura mater and dura mater of the posterior
cranial fossa. Neurology. 1961;11:800-809.
7. Schessel DA, Nedzelski JM, Rowed DW, Feghali JG. Pain after surgery for
acoustic neuroma. Otolaryngol Head Neck Surg. 1992;107(3):424-429.
8. Harner SG, Beatty CW, Ebersold MJ. Impact of cranioplasty on headache after
acoustic neuroma removal. Neurosurgery. 1995;36(6):1097-1100.
9. Schessel DA, Rowed DW, Nedzelski JM, Feghali JG. Postoperative pain fol-
lowing excision of acoustic neuroma by the suboccipital approach: observa-
tions on possible cause and potential amelioration. Am J Otol. 1993;14(5):
491-494.
10. Driscoll CL, Beatty CW. Pain after acoustic neuroma surgery. Otolaryngol Clin N
Am. 1997;30:893-903.
11. Levo H, Blomstedt G, Hirvonen T, Pyykko I. Causes of persistent postoperative
headache after surgery for vestibular schwannoma. Clin Otolaryngol Allied Sci.
2001;26(5):401-406.
12. Silverman DA, Hughes GB, Kinney SE, Lee JH. Technical modifications of
suboccipital craniectomy for prevention of postoperative headache. Skull Base.
2004;14(2):77-84.
13. Jackson CG, McGrew BM, Forest JA, et al. Comparison of postoperative headache
after retrosigmoid approach: vestibular nerve section versus vestibular schwannoma
resection. Am J Otol. 2000;21(3):412-416.
14. Feghali JG, Elowitz EH. Split calvarial graft cranioplasty for the prevention of
headache after retrosigmoid resection of acoustic neuromas. Laryngoscope.
1998;108(10):1450-1452.
15. Soumekh B, Levine SC, Haines SJ. Wulf JA. Retrospective study of post-crani-
otomy headaches in suboccipital approach: diagnosis and management. Am J Otol.
1996;17(4):617-619.
16. Wazen JJ, Sisti M, Lam SM. Cranioplasty in acoustic neuroma surgery. Laryn-
goscope. 2000;110:1294-1297.
17. Sepehrnia B, Knopp U. Osteoplastic lateral suboccipital approach for acoustic
neuroma surgery: technical note. Neurosurgery. 2001;48:229-231.
COMMENTS
E ljamil et al provide an analysis of their experience comparing cra-
niotomy vs craniectomy for the retrosigmoid approach. Although
limited by the retrospective nature of the study, they conclude that
craniotomy is superior with regard to minimizing long-term headaches in
these patients. The authors have taken on an important topic, and they
provide another piece to the puzzle of headaches in these patients.
A prospective randomized trial comparing the 2 modalities of opening
will ultimately be needed to answer the question definitively.
Andrew Thomas Parsa
San Francisco, California
T he authors have provided an insightful study regarding the issue of
postoperative headaches after retrosigmoid craniotomy vs retro-
sigmoid craniectomy. The study summarizes the senior author’s experience
in more than 100 cases in which either craniotomy or craniectomy was
performed. The authors state that as of 2003, their customary approach was
to perform a craniotomy because of an observed lower incidence of
VOLUME 67 | NUMBER 5 | NOVEMBER 2010 | 1291
TEO AND ELJAMEL
cerebrospinal fluid leak. Before 2003, the authors performed both proce-
dures, and these patients have been included in the study as well. Analysis of
various factors associated with postoperative headaches was performed. The
study demonstrated a 43% incidence of headache for the retrosigmoid
craniectomy group vs a 19% incidence of headache for the retrosigmoid
craniotomy group in the initial postoperative period. The incidence de-
creased to 10% vs 1% for craniectomy and craniotomy, respectively, at late
follow-up. The authors conclude that retrosigmoid craniotomy is superior to
craniectomy with regard to postoperative headaches.
1292 | VOLUME 67 | NUMBER 5 | NOVEMBER 2010
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
The paper provides additional support for performing craniotomy
over craniectomy in the posterior fossa. Not only was the incidence of
headache less, but the incidence of cerebrospinal fluid leak was notably
lower in craniotomy group (6.6%) vs the craniectomy group (20%). The
study is limited by small population numbers and lack of randomization;
however, despite these limitations, the article provides an important
contribution to the literature on this problem.
Carlos David
Burlington, Massachusetts
www.neurosurgery-online.com