The results achieved were remarkable. The mortality for small AVMs was between 0 to 5%, and for moderate sized AVMs was generally between 6 and 10%, although some authors found mortality rates of over 20%. Over 60% of patients returned to a full working capacity after operation and serious morbidity was around 10%. Norlén (1949) was particularly successful in that he was able to remove AVMs totally in 10 patients with no mortality and only a small and temporary morbidity. Norlén’s other principal contribution was his statement that “The malformation may cause cerebral circulatory failure. Notice that the arteries of the hemisphere surrounding the AVM, which are hardly seen in the preoperative angiogram, filled normally with contrast once the AVM has been removed. In most cases the postoperative angiograms show that the enlarged and tortuous proximal feeding vessels returned to a normal diameter usually within 2 or 3 weeks.” Following on from this concept, Murphy (1954) first described the concept of “cerebral steal syndrome”.
The First European Congress of Neurological Surgeons (Brussels 1957) included discussion on experience gained in operating on cerebral AVMs. It was generally accepted that palliative procedures such as decompression, ligation of the carotid artery or partial coagulation and partial removal of the lesion were ineffective and that complete removal should be the aim in all possible cases. There remained uncertainty regarding the operability of small or moderate sized lesions in eloquent areas of the brain and in cases of large AVMs. Nevertheless, the first approaches in this direction were already being made by Laine et al. (1956) and Houdart and Le Besnerais who published their results in 1963.
The Limitations of Surgery
Eloquent Areas of the Hemispheres
Pluvinage (1954) predicted a tendency towards more radical removal of all AVMs with the size, shape and location of the lesion becoming a secondary problem.
Tönnis (Tönnis and Schiefer 1959, Tönnis 1961) carried out careful studies of general and localized blood flow and presented 215 patients with cerebral AVMs in which he had achieved complete removal in 118 cases, with 54 AVMs being in eloquent areas. He felt that total removal of an AVM was certainly possible and was the best form of treatment. Preoperative deficits frequently declined after operation and newly acquired deficits were mostly of a temporary nature.
He concluded that
1) The location of an AVM is not a primary reason for inoperability,
2) The preoperative neurological deficits may be reduced after surgery,
3) The mortality in selected cases was 9.5%,
4) Major contraindications to surgery were large voluminous AVMs and elderly patients.
Kunicki and Zoltan described their experience at the 1967 Madrid Meeting of the Congress of European Neurological Surgeons:
Kunicki had successfully removed 2 AVMs from the motor-sensory area and Zoltan described 38 cases of removal of AVMs lying predominantly in the motor cortex or speech areas, in the region of the middle cerebral artery. Of these patients, 4 died postoperatively, and just 2 had mild postoperative neurological disorders. Of 5 patients who had suffered severe hemiplegias following previous hemorrhage 3 were improved after surgery. These authors felt that the main reasons for their success were that the vessels comprising the anomaly did not contribute in any way to the cerebral circulation and that the parenchyma included within and immediately adjacent to the angioma was functionally useless. Zoltan (1968) reemphasized this latter point. Further successes in operating upon AVMs in delicate areas of the brain were presented by Petit-Dutaillis et al. (1953), Laine et al. (1956, 1957), Achslogh et al. (1957), Houdart and Le Besnerais (1963), Pertuiset et al. 1963 and Christensen (1967).
Deep Seated AVMs
Deep seated AVMs lying within the striate, thalamic, parathalamic, limbic, intra- and paraventricular and callosal areas together with most infratentorial AVMs and those within the brain stem had always been generally declared inoperable. However, several surgeons did approach these lesions before microsurgical techniques became available. Olivecrona (1923), David et al. (1934), Alpers and Forster (1945), Boldrey and Miller (1949), Guillaume et al. (1950), Hamby (1952), French and Peyton (1954), Logue and Monckton (1954), McGuire et al. (1954), Carton and Hickey (1955), Strully (1955), Laine et al. (1956, 1957), Leppo et al. (1956), Poppen (1958), Caram et al. (1960), Dereux et al. (1959), Bonnal et al. (1960), Litvak et al. (1960), Pampus et al. (1960), Poppen and Avman (1960), Ralston and Papatheodorou (1960), Odom et al. (1961), Verbiest (1961), Ciminello and Sachs (1962), Levine et al. (1962), Houdart and le Besnerais (1963), Pertuiset et al. (1963), Castellano et al. (1964), Kunc (1965), Mount (1965), Pool and Potts (1965), Laine and Galibert (1966), Walter and Bischof (1966), Lapras et al. (1968), Milhorat (1970), Montant et al. (1971), Fényes et al. (1973), Ribaric (1974).
Kunc (1967) declared that: “For arteriovenous malformations in the basal ganglia and thalamus, ligation of feeding arteries may be the procedure of choice. This, however, carries the danger of producing unintentional infarction owing to the great anatomical variability of the blood vessels at the basal structures of the brain. The same operation can be very successful in one case and produce serious consequences in another. Very good results were achieved with this simple procedure in 2 cases of arteriovenous malformation on the anterior inferior cerebellar artery. To limit the operation to the ligature of supplying arteries is inadequate when the lesion is widespread for the arteriovenous shunt will increasingly attract blood from its small tributaries, which very soon become enormously dilated. Radical removal is the only effective method of treatment, if it is feasible.”
Morello (1967), at the congress in Madrid was of the opinion that “The outlook for patients with angiomas of the basal ganglia is very poor. There are a few accounts of fortunate cases in which the malformation, being small and emerging in the lateral ventricle, could be attacked directly with success, but unfortunately they are often large and cannot be removed.”
Nevertheless, Schürmann and Brock (1967) stated that “The reservations concerning the surgery of AVMs located in vital brain stem centers remain justified. The operability of such lesions seemed to depend upon site, size and clearcut delineation of the angioma, the number, caliber and source of the afferent vessels also whether their origin be uni- or bilateral, and the age of the patient together with the clinical course and picture of the illness.”
In 1967 microtechniques (including the operating microscope, bipolar coagulation, microinstruments and suture material) were introduced and the initial experience in 14 cases (including 4 deep seated AVMs) was published in the monograph of Yaşargil (1969). Splenial and large cerebellar AVMs could be completely removed with good results as presented to the 4th European Congress in Prague 1971.
The Symposium in Giessen (1974) was devoted to the problem of cerebral AVMs and the contributions were published by Pia in 1975. The papers showed a tendency toward more active surgery (Lapras 1975), with the introduction of new techniques such as microtechniques (Pia 1975, Bushe et al. 1975), electrothrombosis (Handa et al. 1975), cryosurgery (Walder 1975) and stereotaxy (Riechert 1975).
The Sixth International Congress of Neurological Surgery in Sao Paulo (1977) dealt once more with deep seated AVMs of the brain. Kunc gave an excellent survey of the achievement and limitations: “It must be recognized that deep seated AVMs are the cause of greater disability and mortality than those at other sites. Hemorrhages threaten function and vitally important structures.”
The large series from the Burdenko Institute was presented by Filatov et al. (1978). In 160 cases the AVM was totally removed, in 60 patients endovascular occlusion of the feeding arteries was performed and 56 other patients underwent various palliative procedures. Of 60 deep seated AVMs, 37 were totally removed and in 16 cases balloon occlusion was performed. There was only one death.
Steiner’s presentation at this meeting, showing results in 35 patients treated with stereotactically directed gamma rays, was a further milestone in the treatment of AVMs. Lesions up to 3 cm in diameter showed startling resolution after such treatment. Another promising radiation technique, especially for large lesions was presented by Kjellberg et al. (1977) in 33 patients.