Lumbar puncture is a simple and harmless procedure. The only danger, that of infection, can be entirely avoided by the exercise of ordinary precautions of asepsis.

The only contraindication is high intracranial pressure. Patients who have brain tumor with signs of increased intracranial pressure, especially choked disc, should be punctured only when this is deemed absolutely necessary for differential diagnosis, and then not more than 2 c.c. of spinal fluid should be withdrawn. Death, caused by hernia of the medulla and midbrain into the foramen magnum, has followed withdrawal of large amounts of fluid in such cases.1

The patient is placed on a convenient table, or a board is inserted under the mattress of his bed. He lies on his side, with the back arched as much as possible and with knees drawn up so that they almost touch his chin. The patient may aid this arching of the back by placing his hands behind the knees and exerting a strong pull. An assistant can keep a restless patient from moving by placing one hand on the nape of his neck and the other behind the knees and thus holding him firmly. Very restless and excited patients must be given a general anaesthetic.

1 Minet and Lavoit. La mort suite de ponction lombaire. L'Echo medical du Nord, Apr. 25, 1909.

Two conditions are essential: the back must not be arched in, but out, and the alignment of the vertebrae must not be scoliotic, but straight. The back is then sterilized with some tincture of iodine, which is removed with a little alcohol. The operator's hands are, of course, also properly sterilized. To mitigate the slight pain incident to piercing the skin, the latter may be anaesthetized with ethyl chloride.

A lumbar puncture needle, sterilized in an oven by dry heat at 150° C. for half an hour, is used. It is best to have several such needles on hand. They can be conveniently placed in cotton-stoppered test-tubes, and if the oven temperature cannot be accurately observed by thermometer it is sufficient to roast them until the cotton begins to turn brown.

This method of sterilization is preferable to boiling the needles, as it is desirable to have them quite dry. Globulin, the detection of which is the object of some of the spinal fluid tests, is precipitated by water. Boiled needles may be used, however, but in that case it is best to discard the first three or four drops of spinal fluid.

The needle should be about 4 1/2 inches long and not larger than gauge 18 nor smaller than gauge 22 of the Brown & Sharpe standard.

The needle is introduced straight into the space between the laminae of the fourth and fifth lumbar vertebrae. This interspace is found by drawing an imaginary line joining the iliac crests. Should this interspace, upon palpation, prove small or narrow, the one above or the one below may be selected instead. The needle is introduced at a point in the midline or a trifle to one side, just below the tip of the corresponding vertebral spine.

Extending from the level of the upper border of the second lumbar vertebra to that of the sacrum is a large meningeal reservoir in which are contained the fibers of the cauda equina. These fibers are loosely held in place and are therefore not injured by the point of the needle. Should the needle touch them, the patient is apt to complain of shooting pains and cramps in the legs. This is no reason for interrupting the procedure. The pain can be eliminated by gently rotating the needle through half a turn.

If in the process of introduction it is felt that the needle is about to strike bone, no attempt should be made to push it further, for then the very sensitive periosteum would be scraped. The operator can easily tell when the needle is about to come in contact with bone, as the resistance of the tendons and ligaments near the vertebrae is greater than that of the more superficial tissues. It is best to withdraw the needle entirely and to try again with another needle.

In some cases it is impossible to get the back of the patient properly arched and aligned. Consequently the projecting spines almost obliterate the small intervertebral spaces. The only possibility of performing lumbar puncture in such a case is by directing the needle at an angle upward. Every puncture should be preceded by a careful palpation of the interspaces. Thus the widest interspace may be selected and the operator must judge, according to the patient's position, at what angle to introduce the needle. The direction of the needle may be changed only after withdrawing it to a level just under the skin, otherwise one runs the risk of impacting it and breaking it off.

A decrease in resistance gives an indication when the meningeal reservoir has been reached and when the stylet is to be withdrawn from the needle. Often the dura gives way with a perceptible pop. A mistake often made is to push the needle too far into the spinal canal; thus the venous plexus at the ventral part of the canal is injured and contamination of the fluid with blood results. Such a specimen can be used only for the Wassermann reaction. It is useless for those tests which presuppose freedom from contamination with such blood constituents as serum albumin and globulin and cellular elements.

Sometimes the needle becomes clogged after some fluid has been collected. In such cases the stylet is reinserted and the needle is turned gently.

About 5 or 6 c.c. of the fluid are collected in a sterile test-tube. It is not advisable to withdraw more for diagnostic purposes, as patients are apt to develop severe headache, faintness, dizziness, or vomiting if too much fluid is withdrawn. Indeed, headache sometimes follows the best technique and greatest care. However, it seldom lasts over a few days.

After lumbar puncture the patient should remain in bed for at least twenty-four hours. Should the above-mentioned symptoms appear and persist, two or three days' rest in bed may be required.