1. The evidence is overwhelmingly in favor of a typical tennis elbow's being caused primarily by a tear between the tendinous origin of the extensor carpi radialis brevis and the periosteum on the anterior surface of the lateral epicondyle. Secondarily, the continual jerks given to this area of acute traumatic inflammation by muscular contractions set up a chronic periostitis here, and to this the symptoms are referable. The acute periostitis occasioned by a direct injury to the epicondyle can become chronic for the same reason, notwithstanding the absence of a tear. Inflammation of the subcutaneous epicondylar bursa and of the radiohumeral bursa is an uncommon but well-defined entity capable of causing the symptoms of a tennis elbow. Ruptures of a muscle belly are very rare and differ neither in pathology nor in treatment from the same condition elsewhere. The possibility of a nipped synovial fringe's being at fault has yet to be proved.
2. The treatment described—deep friction to the tender area, followed by forced adduction of the extended and supinated forearm—has, in the writer's hands, given complete and lasting relief in an average of four treatments (extremes of one and nine), representing a period of eight to fourteen days. Good results followed the treatment of acute and chronic cases (extremes of thirty-six hours and three years), in the old and in the young (extremes of sixty-one and eighteen years), in a case of periostitis visible on the roentgenogram, and in a case due to direct injury to the epicondyle. The method should be adequate in cases of bursitis where bursting is possible; it is unsuited to muscular ruptures which should be treated on general lines,—i.e. by kneading and stretching the belly involved.
3. Mobilization is theoretically justifiable in all cases of typical tennis elbow, because the tendon is thereby pulled off the chronically inflamed epicondylar periosteum to which it is adherent; thus the latter is spared the continual minor traumata which are maintaining the periostitis. Once this is effected, complete symptomatic relief will follow, even though the inflammation at the epicondyle has not yet subsided. Moreover, the writer's treatment is the most convenient from the patient's point of view,—it takes only fifteen minutes every other day; it needs no anaesthetic; and it in no way interferes with the patient's ordinary activities. Furthermore, since healing with permanent lengthening is insured by this method, recurrence is most improbable.
4. Failing mobilization, the best treatment for the recent case appears to be a simple cock-up splint worn on the wrist day and night. In three-quarters of the cases complete relief may be expected in an average of a month's time. The success of this method depends on the relaxation it insures of the extensor carpi radialis brevis, whereby its tendon of origin and the epicondylar periosteum are kept in apposition and allowed to heal together without interruption.
5. All the operations described—on whatever theory they are based—are successful whether or not the elbow joint is opened. Of these, simple division of the origin of the extensor carpi radialis brevis from the bone seems the easiest and the best. Operation is equally indicated in acute and in chronic cases, but appears not to give results superior to the writer's method, since the mere healing of the incision is bound to occupy a fortnight of the patient's time.