directed. An intelligent patient may sometimes be taught to carry out the irrigations for himself. At intervals of not less than twice a week the cavity should be washed out, dried with mops of wool on a probe, and carefully inspected until the edges of the opening have healed and there is no danger of further contraction. If there are signs of contraction Hajek advises that the edges should be cauterized on the tenth day, and again later if necessary. The results of these operations as far as my experience goes are good. The opening remains permanent, and the interior of the sinus soon becomes dry. A cure is usually obtained in two to eight weeks.

Operation with Curette. Occasionally the sphenoidal sinus may be opened in another way. Thus it once happened that in curetting the

FIG. 120. HORIZONTAL SECTION OF THE RIGHT NASAL FOSSA. Showing the relations of the anterior wall of the sphenoidal sinus. 1. Nasal cavity; 2. Posterior ethmoidal cells; 3. Antrum; 4. Sphenoidal sinus.

posterior ethmoidal region for polypi and suppuration the finger, forced well up into the post-nasal space, detected carious bone on the anterior surface of the sphenoid. The softened bone gave way on pressure, and the tip of the finger passed into a large cavity. From its position it was obvious that this cavity must be the sphenoidal sinus: the opening made was therefore enlarged by breaking down the carious anterior wall with the curette, guided by the finger. It may be noted that in this and in another similar case semi-gangrenous polypi associated with an intensely foetid discharge, were present in the nose, and that these polypi rapidly recurred until the curettement had been carried out. The large opening that was made remained permanent, and no after-treatment beyond simple irrigation of the nose was necessary. Three cases treated in this way were all cured,

both as regards the polypi and the suppuration. The method, however, is only applicable when there is extensive caries of the anterior wall of the sinus, and requires general anaesthesia.

Of other methods of operation I have had no personal experience, and must therefore speak more briefly. It has been recommended that when the ostium is invisible, the anterior wall of the sinus should be bored through from the front with an electric drill or burr, or opened with special forceps. Watson Williams has designed a special modification of Grünwald's forceps for this purpose. These methods seem dangerous owing to the great variability in the size of the sphenoidal sinus. I do not think it could ever be justifiable to bore into the anterior surface of the sphenoid in the hope of reaching the sinus unless the position and extent of the latter had been carefully ascertained beforehand by passing a hooked probe into it. Moreover I fail to see how a certain diagnosis can be made except by bringing the ostium of the sinus into direct view by anterior rhinoscopy. If the ostium can be seen the method above recommended can be adopted and offers obvious advantages in its ease and precision.

External Operation. It has also been recommended that the sphenoidal sinus should be reached by an external operation such as I have described for exposure of the ethmoidal cells (see page 333). A curved incision is made round the inner side of the orbit, the ethmoidal cells exposed and cut away one by one until the sphenoidal sinus is reached. This operation is not difficult upon the cadaver, but in the living subject the amount of bleeding that ensues, the small diameter and the great depth of the wound, make it extremely difficult and tedious. The risk of such a procedure seems unjustifiable when there is an alternative, simple, safe and rapid method. Even in acute disease the operation through the anterior nares could probably be carried out with greater speed and precision.

In these circumOnodi examined and found that Under normal con

Jansen's Operation. Lastly, Jansen has recommended that the sphenoidal sinus should be opened through the antrum. Occasionally (see Fig. 120) the sphenoidal sinus comes into close relationship with the maxillary antrum and is separated from it by only a thin plate of bone. stances Jansen's operation may be easily performed. twenty-five skulls with special reference to this point, the operation was anatomically possible in only three. ditions the operator would either first enter the nasal cavity and then the sphenoidal sinus, or he would damage the vessels and nerves which pass through the spheno-palatine foramen. Onodi considers the ordinary nasal route the only safe way to the sphenoidal sinus. It is difficult to see how Jansen's operation could give such a clear view of the field of operation as the methods above described, but it may be a rapid and efficient method of dealing with multiple sinus suppurations, as almost all the sinuses can be attacked at the same sitting. It may also be justified when the nasal fossae are extremely narrow.

To sum up, the sphenoidal sinus may occasionally be opened, whilst operating on the posterior ethmoidal region, by the curette guided by the finger in the post-nasal space, but with this exception, the best plan is to bring the ostium of the sinus into view, and accurately and precisely to remove its anterior wall. This can be done without appreciable risk under cocaine anaesthesia, and with care in the after-treatment a successful result is extremely probable.

The following works may be consulted:

MOURE. Revue hebd. de Laryngol., etc., 1893, xix. p. 817.
LAPERSONNE. (Ref.) Journ. of Laryngol., 1900, xv. p. 49.

HALSTEAD. Archives of Otol., xxx. p. 222.

FURET. Archiv. internat. de Laryngol., etc., 1901, xiv. pp. 1, 181.
GRUNWALD, LACK, etc. Brit. Med. Assoc. Meeting, August, 1902; reported
Journ. of Laryngol., 1902, xvii. p. 598.

M'KEOWN. Lancet, 1902, ii. p. 290.

WRIGHT. Annals of Otol., Rhin. and Laryngol., 1902, Feb.

BERENS. Journ. of Laryngol., 1904, xix. p. 660, and Trans. Amer. Laryngol. Soc., 1904, p. 89.

HAJEK. Archiv für Laryngol., 1904, xvi. p. 105.

GORIS. La Presse Oto-laryngol., Belge, 1903, ii. p. 143.

HINKEL. Trans. Amer. Laryngol. Association, 1902; reported in Laryngoscope,

1902, xii. p. 736.

HOLBROOK CURTIS. Trans. Amer. Laryngol. Soc., 1904, p. 103.

CHAPTER XXIII.

MUCOCELES OF THE ACCESSORY SINUSES OF THE NOSE.

MUCOCELES or bony cysts of the accessory sinuses are by no means rare, although, to judge from the scanty records in medical literature, they have hitherto attracted little attention. The affection is common in connection with the middle turbinate and the ethmoidal cells: more rare in the frontal sinus, whilst the occurrence of mucoceles of the sphenoidal sinus and antrum has not yet been placed beyond doubt.

Definition. A mucocele may be defined as the distension of one of the accessory sinuses of the nose as the result of obstruction of its outlet. The bony cysts in the nose may be classified into three varieties according to their contents. A small number apparently contain nothing but air, the great majority contain mucus or clear fluid and are properly described as mucoceles, while some have muco-purulent or purulent contents. The last are probably the result of infection of a mucocele by pyogenic organisms.

Etiology. The distension of a cell or sinus usually arises from the pressure of retained secretion as the result of temporary, or more usually of permanent, obstruction of its outlet. The obstruction may result from trauma, as Killian believes, but in all probability it is most often due to nasal catarrh. Mucoceles are most common in connection with the ethmoidal cells, which region is most exposed to catarrhal infection: and the cell most often affected, that in the anterior end of the middle turbinate, is especially vulnerable. The frequency with which mucoceles are associated with nasal polypi points to their having a similar origin. When an acute inflammation of the ethmoidal region occurs, when the periosteum becomes thickened and the overlying mucous membrane oedematous, it is obvious how easily the outlets of the ethmoidal cells can become obstructed. If obstruction occur, the fluid secreted by the lining membrane of the cell will accumulate, and cause slow distension of its cavity. If now pyogenic organisms gain admission, the contents of the cavity will become purulent and the lining mucous membrane will be red, swollen, and granulating, or perhaps changed into a pyogenic membrane. Very rarely it seems probable that the opening of the sinus may be congenitally deficient. Thus, I have had under my care a boy, twelve years of age, with bilateral mucocele

of the frontal sinuses. There was no other sign of disease in the nose, and no other cells were affected. The bilateral nature of this affection, and the absence of any sign of disease in the ethmoidal region point to a congenital developmental defect rather than to an inflammatory origin. It is very difficult to understand how trauma could ever cause a mucocele.

Pathology. As already stated, the bony cysts of the nose may contain air, mucus, or pus. The mucus may be clear and yellow; milky or chocolatecoloured; fluid, or containing stringy masses. The cysts which contain air are probably nothing but abnormally large cells. They are lined by normal mucous membrane covered by ciliated epithelium, and the bony wall shows no sign of disease. The inflammatory origin of most mucoceles is shown by the changes which have occurred in the lining membrane. The epithelium consists of one or two layers of cells, the innermost one of which may be ciliated. Beneath the epithelium is a thickened mucous membrane infiltrated with round cells. The periosteum is usually thickened, and in its deeper layers are large multinucleated cells or osteoclasts, in the neighbourhood of which there is usually evidence of rarefying osteitis. Whilst the bone is undergoing absorption on its inner side, fresh bone is being deposited on the outside. In mucoceles of the frontal sinus macroscopic evidence of bone changes is usually present. In parts the walls may have given way altogether, and around the edges of the perforation thickened irregular deposits of bone may be felt, the result of hyperplastic periostitis.

MUCOCELES OF THE ETHMOID.

Bony Cysts of the Middle Turbinate. As already stated, a cell containing air is present in the anterior end of the middle turbinate in about 20 per cent. of skulls, and this cell is very commonly distended. When the cavity contains air only, and the mucous membrane of the middle turbinate is normal, it is probable that even a very large cell is merely a developmental anomaly. They have been regarded as aberrant ethmoidal cells, and may be recognisable in very early life or even in the foetus (Kikuski). A cyst containing mucus or a mucoid fluid arises when the outlet of the cell has become obstructed as the result of nasal catarrh the fluid normally secreted by the mucous membrane accumulates in the cavity and distends it. This is the most common cyst of the middle turbinate and is a true mucocele. Very rarely cysts in this region have been found to contain. pus or muco-pus: this means that subsequent pyogenic infection has taken place.

Symptoms. Cysts of the middle turbinate vary greatly in size; they may give rise to slight enlargement of the anterior end of the bone, or form a tumour completely filling the middle, and even extending into the inferior, meatus. A large cyst may cause considerable pressure on the septum and push it over to the opposite side. The patient complains