Celsus, De Medicina
W. G. Spencer, Ed.

---

4 Again, for fistulae which penetrate so deeply that a medicated bougie cannot be passed down to the ends, or those which are tortuous or multiple, surgery has the advantage over medicine; and there is less trouble if the fistula runs horizontally under the skin, than when it tends directly inwards. Therefore if it lies horizontally under the skin, a probe should be introduced and cut down upon. When there are bends, these are followed up in[p. 309] the same way with the probe and knife; so also when they present multiple branchings. When the end of the fistula is reached, all the callus should be cut out, then pins are inserted through the skin margin, and agglutinating medicaments spread over all. But if it runs straight inwards, after its chief direction has been explored by means of the probe, that cavity ought to be excised, then a pin is to be inserted through the skin opening, and agglutinating medicaments applied as above; or if there is more corrupt ulceration, which is at time the case when there is disease of bone, after the bone has been treated, suppuratives are put on.

Now it is common for fistulae to have their exit between ribs; when this is the case the rib must be cut across on either side at that spot, and the segment removed lest anything diseased be left within. Fistulae which have passed between the ribs often involve the transverse septum separating the viscera above from the intestine. This can be recognized by the position of the fistula and the severity of the pain, and because at times, air with frothy humour escapes from the fistula, especially when the patient has held his breath. In that case there is no opportunity for the medical art. But in the case of other fistulae near the ribs which are curable, greasy medicaments are objectionable but anything else which suits wounds may be used; the best, however, is lint put on dry, or after soaking in honey if anything has to be cleaned.

There is no bone in the abdomen, but all the same fistulae there are so dangerous that Sostratus thought them incurable. Experience, however, shows that this is not always the case. Indeed — and this may[p. 311] seem very remarkable — a fistula which forms over the liver, spleen, or stomach, is safer than one right over the intestine, not because a fistula there is more harmful, but because it opens the way to another danger. Some writers who have had experience of this have shown little perception of the true facts. For often the abdomen is actually penetrated by a weapon, and sutures bring the margins of the wound together and how this is done I will presently point out. Therefore also when a fine fistula breaks through the abdominal wall, it is possible to cut it out, and to join its margins by suture. But if such a fistula widens out inside, this excision necessarily leaves a wide gap which cannot be sutured without applying great force especially in the deeper part where the abdomen is enclosed by a kind of membrane which the Greeks call peritoneum. Therefore, when the patient begins to get up and move about, the sutures break, and intestines prolapse; which causes his death. But these cases are not altogether desperate, and so for the finer fistulae, treatment is to be adopted.

Special consideration is required in the case of those in the anus. In these, where a probe has been passed up to its end, the skin should be cut through, next through this new orifice the probe is to be drawn out, followed by a linen thread which has been passed through the eye made for the purpose in the other end of the probe. Then the two ends of the linen thread are taken and knotted together so as to grip loosely the skin overlying the fistula. The linen thread should be made up of two or three strands of raw flax, twisted up so as to[p. 313] make one. Meanwhile the patient can do his business, walk, bathe, and take food as if in the best of health. Only this thread is to be moved twice a day, but without undoing the knot, the part of the thread outside being drawn within the fistula, and the thread must not be left until it becomes foul, but every third day the knot is to be undone, and to one end that of another fresh thread is tied, and the old thread being withdrawn the new one is to be left in the fistula after being similarly knotted. For thus the thread cuts through the skin overlying the fistula slowly, and whilst the skin released from the thread undergoes healing, that which is still gripped is being cut through. This method of treatment is lengthy but causes no pain. Those in a hurry should constrict the skin with the thread, so that they may continue through more quickly; and at night they should insert into the fistula some fine pledglets of wool, in order that its overlying skin, being put on the stretch, may be thinned out; but these measures cause pain. More speed may be added, but more pain as well, if both the thread and the pledglets are smeared with some one of the medicaments, which I have noted for the eating away of callus. Even here, however, the knife must be used, if the fistula extends inwards, of is multiple. In these kinds of fistulae, therefore, when the probe has been inserted, the skin is to be cut through along two lines so that between them a very fine strip of skin may be taken out, in order that the margins may not unite at once, and that there may be room for the smallest possible quantity of lint to be inserted; all the rest is done in the way described for abscesses. If, however, from one[p. 315] orifice several sinuses lead off, the straight part of the fistula is to be laid open with the scalpel, and the others branching from it, which are now exposed, are to be gripped by a thread. Should any fistula extend so far inwards that it cannot be safely laid open by the knife, a medicated bougie is to be put in. But in all such cases, whether treated surgically or by medicaments, the food should be moist, the drink abundant, and for a while water: when flesh begins to grow up, then at length the patient is to make use of the bath occasionally, and of flesh-making food.