Essential Hypertension Associated with Neurovascular Compression

OI48-396X/92/3006-0834$03.00/0 NEUROSURGERY Copyright

(t) 1992 by the Congress

of NeurologicaJ

B. Kleineberg, Departments of Neurosurgery

M.D.,

Vo!. 30, No.6, !992 Printed in U.S.A.

Surgeons

H. Becker,

l\'I.D.,

Ph.D.,

(BK. ,\fRG. RN) and A.euroradiology

M. R. Gaab, M.D.,

and R. Naraghi,

M.D.

(HB), Hannover Medical School. Hannover, Federal Republic ofGermany

INTRODUCTION Approximately 15% ofall people aged from 20 to 75 years sutTer from hypertension ( II ); therefore it ranks as one of the most frequent diseases in the Western hemisphere. Even after extensive pathophysiological investigation. the pathogenesis remains unclear in 90 to 95% of patients and is described as ..essential" hypertension. According to the World Health Organization, hypertension exists if the blood pressure exceeds 160/95 mm Hg (49) and often leads to lethaI complications if untreated (3). Cerebral hemorrhage and myocardial infarction are the most frequent causes of death. The etiology of essential hypertension may be multifactorial. High salt consumption. obesity, alcohol, and a genetic predisposition are some of the known factors contributing to the development othypertension. A possible relationship between central nervous system disorders and essential hypertension. however, has also been discussed (6. 13,34). Since 1973, Jannetta et al. (20) have thought that essential hypertension might be associated with vascularcompression of the root entry zone (REZ) of cranial nerves IX and X byectatic loops of posterior fossa arteries. Most of the hypertensive patients treated with microvasculardecompression ofthe REZ of left cranial nerves IX and X showed a significant decrease in blood pressure postoperatively and at a follow-up period of at least 6 months ( 19. 20). The blood pressure remained unchanged in patients having surgery on the right side. These results were surgically confirmed by Fein et al. ( 13. 14). Van Ouwerkerk et al. (46), and anatomically by Naraghi et al. (30). A primate model of Jannetta et al. (21) and Segal et al. (39.40) provided further substantiation. This prompted us to investigate the relationship between hypertension and microvascular findings by retrospective evaluation of angiograms in hypertensive patients. In addition. we intended to provide anatomical and radiographic criteria to predict a vascular compression preoperatively in patients with primary hypertension (23).

Ph.D.,

MA TERIALS Radiographic

localization

AND

METHODS

qr the root entry

zone

The roots of left cranial nerves IX and X were marked by arterial clips in 10 cadavers, similar to the method of De Lange et al. (9). Radiographs in standard projections then provided information on the position of the REZ of cranial nerves IX and X. A left retromastoid craniectomy was performed on each cadaver with the head fixed in a Mayfield clamp in the prone position. The left cerebellar hemisphere was caretully retracted using an operating microscope. After preparation ofthejugular foramen. cranial nerves IX and X were identified. Three small clips were placed on the nerves. one proximal to the brain stern (Fig. 1). This served to mark the REZ of left cranial nerves IX and X. Finally, the dura and galea were replaced and sewed into the

FIG. I. Anatomical specimen ofthe posterior fossa on the left side. dorsal view. The glossopharyngeal and vagus nerves are marked by three clips. The proximal clip (arro»'j represents the REZ. PICA. posterior inferior cerebellar artery; OL. olive; IX. g!ossopharyngeal nerve; X. vagus nerve: XI. accessory nerve.

June 1992

NEUROV ASCULAR

original position. Radiographs were then taken in lateral. Towne, and anteroposterior (AP) projections (Fig. 2, .4-C). The cadavers had normal brains, proven by neuropathological section in each ofthe cases. Five ofthe cadavers examined were male and five were fernale, between the ages of 40 and 74 years (mean, 56.2 yr); this corresponded to the ages of hypertensive patients whose angiograms were studied. Angiograms

and eva'uation;

hypertensive

and normotensive

patients All venebral angiograms performed at the Hannover Medical School from 1981 to 1988 were evaluated (n = 1238). Of930 available patient histories, 107 were identified as patients with definite essential hypenension; that is. the patient had antihypenensive treatment for at

COMPRESSION

AND

HYPERTENSION

835

least 2 years, blood pressure was higher than 160/95 in more than 10 measurements, and, if secondary hypertension was excluded, essential hypertension was supposed, To provide data comparable with cadaver findings, only patients from 40 to 74 years old were studied (mean, 57.0 yr); 64 were women and 43 were men. As a control group, 100 angiograms ofnormotensive patients were evaluated for comparison (age range, 40 to 73 yr; mean, 55.2 yr; 48 women and 52 men). The indications for angiography in hypertensive and normotensive patients was subarachnoidal hematoma, arteriovenous malformations, trigeminal or glossopharyngeal neuralgia, facial spasm, and some small tumors of the convexity. Three angiograms from patients with proven renal hypertension were identified and were analyzed as weil. Comparison

of arter.v diameters

To obtain infonnation on whether the diameter ofthe arteries has any influence on the probability ofhypertension, th~ diameter ofthe contrast-filled lumina of every artery in the REZ of left cranial nerves IX and X was measured in the hypertensive and nonnotensive patients. To obtain comparable results, all measurements were perfonned in the Towne projection. The magnification factor was detennined to be exactly 0.84 (source-to-object distance, 84 cm; sourceto-film distance, 100 cm). After marking the contrast-filled lumen with a lead pencil, measurements were taken by using a transparent ruler to an attempted accuracy ofO.1 mm, according to the procedure ofGabrielsen and Greitz (16).

RESULTS

FIG. 2. Radiographs ofmarked cranial nerves IX and X in Towne (.4 ), AP (B), and lateral ( C) projection. The proximal clips (arrow.\") represent the REZ of left craniaJ nerves IX and X.

The proximal areas of the superimposed clips of all patients in one projection described the REZ of left cranial nerves IX and X. In each projection, areas of only 50 to 59 mm2 resulted (Fig. 3, A-C). The clips were projected 10 mm to the left ofthe skull midline, referred to the geometric center of the area, and exactly horizontally to the inferior border ofthe intemal acoustic meatus in the Towne projection. The area ofthe clip markings in the AP projection lay 10 mm to the left ofthe midline and 5 mm caudally to the inferior border ofthe intemal acoustic meatus. In the lateral projection, the clips appeared 20 mm caudally to the petrous ridge and 15 mm dorsally to the clivus. From these anatomical findings, overlay patterns (Fig. 4) were developed, which included all three projections-Towne. AP, and lateral. The angiograms could be easily evaluated using these patterns. An artery had to appear in the REZ in at least two projections (lateral and one other projection. Towne or AP) to be labeled as crossing the REZ. An artery crossed the REZ in 68 of the hypertensive patients. whereas 17 angiograms showed no apparent artery in the REZ. Because of thick bone in the lateral projection or unsatisfactory technique. 22 angiograms could not be evaluated. The angiograms that were evaluated showed 80.0% with an artery in the REZ and 20.0% without a distinct artery for the hypertensive patients (Fig. 5). The posterior inferior cerebellar artery (PICA) appeared with highest frequency in the REZ (24 cases). In 20 angiograms. the vertebral artery (V A) crossed the REZ, and in 9 angiograms both the PICA and the V A crossed the REZ. In 13 cases, the anterior inferior cerebellar artery (AICA) was observed in the REZ, in 1 angiogram the basilar artery (BA) was observed. and in 1 case the V A and AICA were observed (Fig. 6). Only 29 ofthe 100 angiograms ofnormotensive patients (control group) showed an artery in the REZ. In 55 cases. no artery crossed the REZ. and 16 angiograms could not be evaluated. Again referring to the angiograms that could be evaluated. 34.5% showed an artery in the REZ and 65.5% had no artery in the REZ (Fig. 5). In 13 cases. the V A, in 10 cases, the AICA. in 5 the PICA, and in 1 case. the V A and PICA crossed the REZ (Fig. 6). In 2 cases. the angiograms of the 3 renal hypertensive patients showed no artery in the REZ. and in 1 case. the

836

KLEINEBERG

et al.

1'0/.30. No.6

Neurosurger}

A

AP Projection

Towne Projection

FIG. 4. Overlay pattern developed from superimposing the radiographsofmarked cranial nerves IX and X (Fig. 2), including Towne. AP. and lateral projections. HYPERTENSION

NORMOTENSION

c \ I

~.;:.Arterv FIG.

55 {N=I001

IN~107) in REZ

5. Frequency

-= No Artery in REZ

of arteries

crossing

the

hypertensive and normotensive patients. The grams that could not be evaluated is shown.

~

f

FIG. 3. Superimposed clips of all 10 cadavers in Towne (A ), AP (8), and lateral (C) projection. The REZ is represented by the proximal clips (arrow). AICA crossed it. The differences between normotensive and hypertensive patients are statistically significant based on a x2 test for variation (x2 = 35.73. P