As a result of prolonged inflammation edematous polypi and granulomata are not infrequently seen, but true benign tumors of the esophagus are rare affections. Keloidal changes in scar tissue may occur. Cases of retention, epithelial and dermoid cysts have been observed; and there are isolated reports of the finding of papillomata, fibromata, lipomata, myomata and adenomata. The removal of these is readily accomplished with the tissue forceps (Fig. 28), if the growths are small and projecting into the esophageal lumen. The determination of the advisability of the removal of keloidal scars would require careful consideration of the particular case, and the same may be said of very large growths of any kind. The extreme thinness of the esophageal walls must be always in the mind of the esophagoscopist if he would avoid disaster. [210] CHAPTER XXV--ENDOSCOPY IN MALIGNANT DISEASE OF THE LARYNX The general surgical rule applying to individuals past middle life, that benign growths exposed to irritation should be removed, probably applies to the larynx as well as to any other epithelialized structure. The facility, accuracy and thoroughness afforded by skilled, direct, laryngeal operation offers a means of lessening the incidence of cancer. To a much greater extent the facility, accuracy, and thoroughness contribute to the cure of cancer by establishing the necessary early diagnosis. Well-planned, careful, external operation (laryngofissure) followed by painstaking after-care is the only absolute cure so far known for malignant neoplasms of the larynx; and it is a cure only in those intrinsic cases in which the growth is small, and is located in the anterior two-thirds of the intrinsic area. By limiting operations strictly to this class of case, eighty-five per cent of cures may be obtained.* In determining the nature of the growth and its operability the limits of the usefulness of direct endoscopy are reached. It is very unwise to attempt the extirpation of intrinsic laryngeal malignancy by the endoscopic method, for the reason that the full extent of the growth cannot be appreciated when viewed only from above, and the necessary radical removal cannot be accurately or completely accomplished. * The author's results in laryngofissure have recently fallen to 79 per cent of relative cures by thyrochondrotomy. _Malignant disease of the epiglottis_, in those rare cases where the lesion is strictly limited to the tip is, however, an exception. If amputation of the epiglottis will give a sufficiently wide removal, this may be done en masse with a heavy snare, and has resulted in complete cure. Very small growths may be removed sufficiently widely with the punch forceps (Fig. 33); but piece meal removal of malignancy is to be avoided.