Keywords
pituitary apoplexy - oculomotor nerve - anatomical variations - sella turcica
Introduction
Pituitary apoplexy is caused by an acute ischemic stroke, with secondary bleeding
after the stroke, or it could be due to a primary bleeding event in pituitary adenomas.
Some evidence indicates that pituitary apoplexy is associated with the sudden onset
of neurological and/or hormonal disorder.[1] One possible neurologic symptom is paresis/plegia of the third cranial nerve. Lesions
of the third cranial nerve are rare. However, the possible cause of a lesion of the
oculomotor nerves passing through the cavernous sinus as well as the trochlear and
abducens nerves and visual acuity deficits may be an acute onset of pressure on the
cavernous sinus. However, the most common nerve lesion is the oculomor nerve with
varying degrees of paresis.[2] The nerve lesion of the oculomotor nerve is very rare (14.28% from all pituitary
apoplexy), though the incidence in the patient group is relatively high and the formation
of this nerve lesion has always been associated with pituitary tumor apoplexy (100%).[3] Possible causes of the nerve lesion and anatomical variations are discussed by the
authors.
Materials and Methods
From April 2011 to August 2012, 129 patients underwent surgery for pituitary adenoma
at the Department of Neurosurgery, First Medical Faculty and Central Military Hospital,
Military Faculty Hospital, Prague. The group of patients comprised 62 women and 67
men, with a mean age of 52.5 years (range: 9–81 years). In the patient group, there
were 51 hormonally active tumors and 78 were hormonally inactive (requiring surgery
for visual disorder or for compression of the surrounding structure). Four patients
(3.1%) presented lesions of the oculomotorius nerve—complete oculomotor nerve palsy
(ptosis, diplopia, and mydriasis). In this group of four patients (mean age: 55 years,
range: 36–65 years), all patients (two women and two men) underwent surgery. Early
clinical symptoms appeared anywhere from 1 to 3 weeks before the surgery. The appearance
of lesions of the oculomotor nerve was sudden in all cases. One patient suffered from
panhypopituitarism preoperatively. The tumor size varied from 18 to 23 mm (diameter
20 mm). In one case, a tumor penetration into the cavernous sinus was demonstrated
based on preoperative magnetic resonance imaging (MRI). Disintegrating portion of
tumor and marks of bleeding were typically presented on the MRIs. In two patients,
the arrested pneumatization and thickened bone were observed on MRI. In the remaining
two patients, a thickened diaphragm was more easily noted after the _removal of the
tumor in the control MRI scan the next day.
Since November 2006, our team has used the binostril endoscopic endonasal approach
with the aid of a Storz 0 degree endoscope (Karl Storz Endoscope). The binostril technique
(four-hand technique) is the standard surgical technique in our department. This procedure
does not require septum lateralization. We resect only a small area (0.5 × 0.5 cm)
in the posterior part of the nasal septum. The use of both nostrils offers several
advantages, including greater working space and a large range of instrument movement.
At a certain point during the surgical procedure, two instruments have to be used
simultaneously, namely, a drain cannula and an endoscope (hence, the four-hand technique).
Our operations take place in a multifunction room, where data are transmitted from
an intraoperative 3.0-TMR system (General Electric) and subsequently updated in a
surgical navigation system (Brainlab).
Results
In two cases, there was the absence of the sphenoid cavity, as well as evidence of
thickened bone. In these two cases, it was necessary to grind the bone (width from
6 mm to 1 cm) ([Fig. 1]). A very solid opaque diaphragm was macroscopically visible after the removal of
the tumor and thickened bone in the remaining two cases ([Fig. 2]). The diaphragm was whitish, smooth, and strong in appearance. There was no typical
decline of the diaphragm in these two cases after the removal of the tumor. A tougher,
but fragile, bruising tumor with initial high pressure was found in all cases after
cutting the dura mater. After cutting the dura mater, there was always a massive sharp
leakage of old blood as a part of the sellar expansion. All patients showed a very
fast recovery. Partial recovery generally occurred during the first 24 hours. Complete
adjustment was observed in all patients within 1 week after the surgery. MR radical
resection (3 months postoperatively) and hormonal radical resection (for hormonal
active tumors) were achieved in 73% of the cases with the help of intraoperative MRI
(129 operations for pituitary adenomas in all patients). Magnetic resonance (MR) radical
resection of a subgroup of patients (the subgroup consisted of adenomas with lesions
of the oculomotor nerve) was achieved in three adenomas, 75%. On postoperative MRI,
we observed the remains of the left side of the adenomas in the lateral compartment
of the cavernous sinus. We did not encounter any postoperative complications in the
cases with lesions of the oculomotor nerve. Pituitary hypofunction remained in patients
with substituted hypofunctional hypophysis before surgery. Postoperative substitution
has been necessary in two patients, 5 years after the surgery. Histologically, bruising
was seen in all of the adenomas.
Fig. 1 Arrested pneumatization, as well as evidence of thickened bone on magnetic resonance
image.
Fig. 2 Very solid opaque diaphragm on magnetic resonance image.
Discussion
Pituitary apoplexy is caused by acute ischemic injury or spontaneous hemorrhage of
the pituitary tumors. The frequency of stroke ranges from 2 to 7%.[4] However, it is important to determine whether it is a clean bleeding, which is shown
on MR, or clinically developed clinical symptoms (sudden malfunction of the pituitary
gland, visual disturbances, or disorders of the ocular nerves). The increased fragility
of tumor vessels has been suggested as a cause of pituitary apoplexy.[1] The embolism in atherosclerotic disease has been also suggested as a cause of pituitary
apoplexy.[4] Predisposing factors of pituitary apoplexy include high blood pressure, head injury,
pituitary irradiation in the past, anticoagulant therapy, treatment with dopamine
agonists and pregnancy, diabetes mellitus, high altitude, dengue hemorrhagic fever,
dynamic endocrine tests, and spinal anesthesia.[5]
[6] The latter factor can also support the hypothesis that the rapid enlargement of
pituitary adenoma may lead to the oppression of veins, which are a part of the portal
system in the pituitary stalk through diaphragm. This condition can lead to congestion
and hemorrhage[7]; however, small tumors can also develop apoplexy events.[3] Intratumoral vasculopathy rendering the vessels more susceptible to hemorrhage has
been proposed as an option.[3] Furthermore, a significant correlation between the intratumoral expression of vascular
endothelial growth factor and tumor necrosis factor-α and the presence of tumor hemorrhage
has been found, suggesting a possible causal relationship, although other group did
not confirm this finding.[3]
Lesions of the oculomotor nerve that result from pituitary apoplexy can have several
causes. The notion of direct pressure on a nerve by a large pituitary adenoma is apparently
the most logical cause.[8]
[9] Pituitary apoplexy is an acute condition in which the wall of the cavernous sinus
is not primarily infiltrated yet is enhanced pressure inside the sinus. The conjecture
pertaining to vascular lesions of the oculomotor nerve, which is based on a reflex
spasm of vessels from the pressure of the tumor that encases the internal carotid
artery, is much more intriguing.[10] This belief explains the emergence of acute lesions, but cannot explain why it would
be limited only to the disability of the oculomotor nerve and not to the trochlear
and abducens nerves.[11] Presently, the theory that is best is still being debated. The basic premise is
that the oculomotor nerve, which passes through the lateral wall of the cavernous
sinus, is much more vulnerable than the other nerves. It has been suggested that the
abducens nerve is more vulnerable than the oculomotor nerve. The cause of the sixth
cranial nerve lesion is beyond the cavernous sinus and this lesion is described more
often in relation to either an inflammation or to a skull base tumor.[12] The highest position of the third cranial nerve (the oculomotor nerve) in cavernous
sinus is another explanation as to why the nerve is affected. The anterior part of
the cavernous sinus is relatively the most mobile part.[13]
We believe that the most likely cause of the disability of the oculomotor nerve is
the anatomy of the sella turcica and its surroundings. The frontal, caudal, and dorsal
surfaces are made up of bone. The cranial saddle surfaces are covered with the diaphragm.
The lateral surface forms the inner walls of the cavernous sinus. The pressure is
transmitted to the wall of the cavernous sinus if the diaphragm is very rigid, or
if the base of the saddle is rigid. The cavernous sinus is not an open tube resembling
an enlarged vein, but rather it is a patchwork of lagoons (sometimes described as
spicules, which form a dense network). This structure resembles a sponge, with high
blood flow.[14] Congestion in the sella turcica leads to high pressure inside the cavernous sinus.
The oculomotor nerve is located uppermost, above the trochlear nerve in the lateral
wall of the cavernous sinus at the transition to the upper surface. These conditions
explain why the highest pressure is at this particular point. The conditions also
explain why the site of the pressure is at the lesion of the oculomotor nerve.
This hypothesis is confirmed by our results. All operations were performed by the
four-hand endoscopic technique through two nostrils (standard procedure at our institution).[15] In all patients, perioperative MRI was applied at the same time. The maximum radicality
of tumor removal is guaranteed by this technique at this time.[16] No complications were seen in any of our patients. All patients were transmitted
by car to the Department of Endocrinology at the Charles University Hospital in Prague.
As our subgroup consisted of only four patients, no statistical analyses could be
done. Nevertheless, we found some evidence to support our hypothesis. The most important
finding concerns the underdevelopment of the sphenoid sinus with a solid saddle base
that was seen in two patients (conchal type of sphenoid sinus). In the remaining two
patients, we observed a very strong diaphragm that became visible through MRI after
tumor removal and thickened bone (presellar type of sphenoid sinus). We found anatomical
variety in all patients with lesions of the third carinal nerve. This observation
is a new concept which was found. These findings suggest that the primary pressure
is at the lateral wall of the cavernous sinus. The incidence of presellar type was
reported to be 5.5 to 27% in the literature and conchal type is reported to be none
to 3%.[17]
[18]
[19]
In the whole group of 129 patients (including hormonally active microadenomas), the
average tumor size was 21 mm; in the subgroup (lesions of the oculomotorius nerve),
tumor size ranged from 18 to 23 mm (diameter 20 mm). These findings suggest that lesion
formation is not determined by the size of the tumor, but rather by the speed of the
pressure buildup in the event that anatomical variations are present where the pressure
is directed toward the cavernous sinus.
The hypothesis that pressure is the cause of the oculomotor nerve lesion was confirmed
by the rapid speed of adjustment, that is, all patients demonstrated partial modification
in the first 24 hours after the surgery. Moreover, all patients were able to lift
their eyelids on the first postoperative day.
Conclusion
Early tumor resection is the treatment of choice for lesions of the oculomotor nerve,
which are caused by pituitary apoplexy. The hypothesis that sudden pressure causes
oculomotor nerve lesions is highly plausible in the event that anatomical variations
are present, but this could not be demonstrated statistically in the present report
because of the small number of cases.
