duced to 75 mg. per day. After three months of therapy, the patient voluntarily stopped treatment with phenformin. Comment The use of any drug requires that the physician be thoroughly informed as to both its capabilities and its limitations. To date, tissue toxicity has not been reported with use of the hypoglycemic biguanide compounds, after careful evaluation of some 3,000 patients over a period of three years." Toxicity, therefore, is not a limitation of phenformin. In this series, no evidence of toxicity has been found. A limitation in the use of phenformin has been the occasional development of acetonuria in the presence of aglycosuria and lowered blood sugar levels, indicating a starvation type of ketosis without acidosis. Some patients in this series, quite early in their treatment, did indeed show ketosis without glycosuria. In a few instances this was accompanied by slight weight loss, even though the patient's control had to be considered excellent according to blood sugar levels and urine sugar values. In every instance, this problem was easily met by a moderate increase in the carbohydrate content of the diet, plus a slight reduction in the dose of phenformin, usually 25 mg. per day. Another limitation has been the occurrence of gastrointestinal side-effects. Although the problem of side-effects with phenformin therapy has been widely discussed,' gastrointestinal reaction serious enough to warrant discontinuance of this drug occurred in only 5 out of 104 patients in this series. A somewhat larger number of the patients studied had slight gastrointestinal distress, nausea, or frequent bowel movements when first started on the drug, but were encouraged to continue therapy. In spite of these initial gastrointestinal symptoms, as they continued to take the drug, the side-effects became less and less severe and finally disappeared altogether. Three patients in whom side-effects continued after six weeks of therapy found that they were able to limit the side-effects by taking rest periods from the drug of 24 hours' duration every two weeks. In no instance did nausea and vomiting ever become so severe as to produce keto-acidosis, even in those patients in whom the therapy later was withdrawn. Seven months' experience with phenformin therapy in 104 patients of widely varying ages, degrees of severity of the disease, economic backgrounds, and degrees of emotional adjustment to diabetes is presented in this study. The breadth of usefulness of phenformin in the management of diabetes is impressive. Patients from the ages of 3 years to 83 years have been well regulated with the drug. Patients whose insulin requirements have been 700 units a day and those who have never received insulin therapy have shown from good to excellent response to this drug. In many instances, phenformin has been responsible for adjusting life situations so that patients whose livelihood was threatened, whose peace of mind was disturbed because of marked lability of their diabetes, have been restored to more useful and certainly more serene lives. Too often, one hears both patients and physicians emphasize the abolition of insulin from the management of the diabetic as a criterion for the effectiveness of an orally given hypoglycemic agent. Insulin must still be considered the most proved and effective therapy for diabetes. However, most clinicians are aware of the many gaps in the effective control of diabetes which insulin leaves because of its very nature. In an effort to smooth the course of management and enhance the control of the problem patient, phenformin has been used as adjuvant to insulin in this study. When it was possible to do away with insulin, this was considered a happy circumstance, but it was not necessarily the primary objective. Phenformin has been used by others with success as complete insulin replacement therapy in patients with mild, maturity-onset diabetes. However, one of the primary efforts in this study had been to ascertain the value of the drug in controlling those patients with problem diabetes for whom the use of oral therapy would not preIviously have been considered. Many patients in this study were started on phenformin therapy because of recurrent hypogly. cemic episodes which they had suffered, in some cases for several years prior to phenformin treatment. These patients probably represented the most difficult group in the entire study. In every case but one, it was possible, after the institution of phenformin treatment, to eliminate these hazardous episodes of hypoglycemia while at the same time improving the level of diabetic control. This in itself has represented one of the most gratifying aspects of this drug. A patient, aged 52, had for several years prior to taking phenformin suffered from severe episodes of hypoglycemia which came on without warning symptoms. Eighteen months prior to beginning phenformin therapy, this patient had been treated with intravenously given glucose seven times for severe convulsive attacks of hypoglycemia. In spite of the patient's best efforts, and frequent adjustments of insulin dosage, the patient's diabetes was considered poorly controlled. Since beginning phenformin therapy, his insulin requirement dropped only 12 units, from 52 units a day to 40 units a day. Control, however, has been remarkably improved, with long periods of aglycosuria, much more satisfactory postprandial blood sugar levels, but, most importantly, complete freedom from insulin hypoglycemia. A 37-year-old patient with diabetes since the age of 13 had likewise been plagued for several years with severe and frequent episodes of hypoglycemia which occurred without warning symptoms. The patient had been warned by his employer that he could not be continued as an employee, if further reactions occurred, because of the hazard to the patient himself and other employees. Since starting on phenformin therapy, the patient's insulin requirement has dropped from 68 units of insulin a day to 56 units of insulin a day. Control has improved remarkably, and the patient has not had a single further episode of severe insulin reaction in seven months at the time of this writing. Freedom from hypoglycemic reaction has been considered a real benefit derived from the use of phenformin therapy," and this has indeed been my experience. It should be noted that, early in the course of treatment with phenformin, in patients whose insulin requirements have been high or in those who have a past history of insulin sensitivity, hypoglycemia with symptoms may oc cur. In every instance in this series this was due to failure to reduce the insulin dosage rapidly enough after the inception of phenformin treatment. In only one instance was an episode of hypoglycemic reaction seen in a patient after he had received phenformin for more than one month. In the early adjustment period it is, therefore, urged that the physician be on the alert for hypoglycemia in the patient and prepared to adjust insulin requirements quickly in the insulin-sensitive patient or in those who have previously had an insulin requirement in excess of 80 units per day. In the evaluation of the patients in this study, it is impressive how many patients were responsive to the hypoglycemic effects of phenformin. Ninetysix patients or 92% showed at least some measure of responsiveness to the drug in the dosage levels utilized. Only eight patients showed no hypoglycemic effects from use of the drug in doses of 150 mg. The five other patients who were classified as failures showed good to excellent responses, but the drug had to be withdrawn because the patients were unable to tolerate the gastrointestinal side-effects. This, therefore, gives a response rate to phenformin of 92% in this series. Of the entire series, 51%, or 53 patients, had excellent results according to the criteria described previously. Twenty-nine patients, or 28%, had good results, representing a good to excellent response rate in this series of 79%; nine patients had only fair results. This compares favorably with studies done on the effectiveness of other orally given hypoglycemic agents, with the notable exception that most patients in this series would not be considered favorable candidates for successful oral therapy with sulfonylureas. Of the 53 patients who are considered to have excellent results, 19 had previously been poorly controlled on various therapies. Only four had previously been considered excellently controlled. Of those having a good response to phenformin, 21 had previously been poorly controlled, and only 8 had been satisfactorily controlled prior to the institution of phenformin therapy. Among those in whom the drug was withdrawn because of gastrointestinal side-effects, one had good control and one had fair control; however, none had achieved excellent control with other therapies. The patients who were unresponsive to administration of the drug had also not responded well to other techniques employed in the management of their diabetes. Only six patients in this series had previously taken other orally given hypoglycemic agents prior to phenformin. Two patients who had been classified as failures on tolbutamide therapy, requiring a return to insulin one year before phenformin became available, were both excellently controlled with the use of phenformin; one with phenformin in combination with insulin and one with phenformin alone. Two patients treated successfully with chlorpropamide immedi ately prior to phenformin, responded excellently to the phenformin therapy. There were two patients on tolbutamide therapy, with poor control when first seen; one had an excellent response to phenformin without insulin and one a good result with insulin and phenformin. Table 3 shows the response to phenformin therapy according to the duration of diabetes. In summary, 83 patients did significantly better with phenformin alone or phenformin plus insulin than they did on the regimens used prior to this study. An analysis of the failures reveals the following pertinent data. The greatest number of persons classified as failures originate in the labile maturity-onset group and in the growth-onset group who had passed the age of 15 years. There were only two failures among the stable maturity-onset group. It was quite interesting that both patients responded well to the phenformin, but the drug had to be withdrawn because of gastrointestinal side-effects. In the maturity-onset labile group, six failures were reported. In this group, there were three failures due to lack of responsiveness and three failures because of gastrointestinal sideeffects. In patients in the growth-onset group whose ages were still under 15, there were three failures. In the growth-onset type diabetics, the degree of responsiveness to the drug must be correlated with the degree of cooperation that one can expect, both from the patient and the parents of this type patient. It would be interesting to speculate on the degree of response which might be seen could one obtain ideal cooperation in those cases. Responsiveness to phenformin as to other orally given hypoglycemic agents seems, therefore, to be best in those patients who have had diabetes less than 20 years, whose onset was after the age of 30, and whose diabetes had been relatively stable on prior therapies. However, it is important to note that, in the entire series, only 44 patients were considered satisfactorily controlled prior to phenformin therapy. Of the 44 patients, only 3 were controlled less satisfactorily after the institution of phenformin therapy, and of these 2 were frank failures. Fifty-one patients were considered poorly controlled prior to treatment with phenformin. In forty-two of these patients, control improved from fair to excellent. By far the largest number of failures occurred in previously poorly controlled diabetics on other therapies. The remaining nine patients in the series were newly diagnosed diabetics. In this group, there were seven excellent responses and two frank failures. In those patients in whom it was possible to treat the diabetes with phenformin alone, nine had never received insulin: five had complete replacement of insulin with phenformin; and, in three cases, it was possible to switch the patient 3 from treatment with sulfonylureas to phenformin Other than gastrointestinal side-effects, which necessitated the withdrawal of phenformin in five patients in this series, failures with phenformin therapy might be ascribed to the following problems: (1) emotional problems, (2) dietary indiscretions, and (3) lack of physiological response to the hypoglycemic action of the medicament at the dosage levels considered practical for this study (150 mg/day). Other investigators have reported good tolerance and satisfactory control of many diabetic patients with daily phenformin doses of 200 to 400 mg.' It is possible that some of the eight patients who failed to respond to the drug because of the daily maximum dose limit of 150 mg. might have done well on higher doses. This experience shows, however, that with the judicious use of phenformin in moderate dosage, a high rate of hypoglycemic response may be achieved. Summary and Conclusions The influence of a new non-sulfonylurea orally given hypoglycemic agent, phenformin hydrochloride (DBI), on the control of the diabetes in a group of diabetic patients was evaluated. These patients were selected for study because they presented problems in diabetes management. The study covers a period of seven months. In the dosage range of 37.5 to 150.0 mg. per day, in divided doses with meals, phenformin with or without adjunctive exogenous insulin made it possible to achieve fair to excellent control of the diabetes in 91 of these 104 patients (88%). Eighty-three (80%) of 104 patients experienced significant improvement in diabetic control after the use of phenformin. Eight patients (8%) failed to be controlled with phenformin therapy because of behavioral or environmental problems, dietary abuse, or lack of adequate response at dosage levels used. Another five patients (4%) failed because gastrointestinal symptoms necessitated withdrawal of administration of the drug. Phenformin is a safe and effective orally given hypoglycemic drug which is sufficiently wide in its range of activity to be considered the oral therapy of choice in the management of diabetes. 449 W. Tenth St. The phenformin was furnished as DBI for this study by the U. S. Vitamin & Pharmaceutical Company, New York City. References 1. (a) Sugar, S. J. N., Thomas, L. J., Eugenio, T. M.. and Tatlier, S.: Phenformin (DBI) in the Management of Diabetes Mellitus, M. Ann. District of Columbia, 28:426431 (Aug.) 1959. (b) Pomeranze, J.; Mouratoff, G. T.; Gadek, R. J.; and King, E. J.: Phenethylbiguanide, New Orally Given Hypoglycemic Agent: Report After Two Years of Clinical Experience, J. A. M. A. 1711252-258 (Sept. 19) 1959. (c) Krall, L. P., and Bradley, R. F.: Clinical Evaluation of Formamidinyliminourea, New Biguanide Oral Blood Sugar Lowering Compound: Comparison with Other Hypoglycemic Agents, Ann. Int. Med. 501586-613 (March) 1959. 2. (a) Pomeranze and others. (b) Krall and Bradley." (c) Skillman, T. G.; Kruger, F. A., and Hamwi, G. J.: Metabolic and Endocrine Studies with Phenethyldiguanide (DBI), Diabetes 81274-278 (July-Aug.) 1959. 3. Freedman, L., Ungar, G.; Shapiro, S. L.; and Sadow, H. S.: Phenformin (DBI), Pharmacology, Toxicology and Clinical Use in Diabetes, A. M. A. Scientific Exhibit, Atlantic City, N. J., June 8-12, 1959. 4. (a) Pomeranze and others. (b) Krall, L. P., White, P., and Bradley, R. F.: Clinical Use of Biguanides and Their Role in Stabilizing Juvenile-type Diabetes, Diabetes 7:468477 (Nov.-Dec.) 1958. 5. (a) Sugar and others. (b) Pomeranze and others. (c) Krall and Bradley, (d) Krall and others. (e) Reinikainen, M.: Phenethyldiguanide for Diabetes, Ann. Med. Int. Fenniae 48:25-31 (Jan.) 1959. (f) Skillman and others. (g) Freedman and others." (h) Krall, L. P., and Camerini-Davalos, R.: Clinical Trials with DBI, New Nonsulfonylurea Oral Hypoglycemic Agent, A. M. A. Arch. Int. Med. 102:25-31 (July) 1958. 6. (a) Krall and others. (b) Reinikainen. 7. (a) Krall and Bradley. Davalos." (b) Krall and Camerini |