راق

H

YPOGLYCEMIG agents have been used with varying degrees of success in the management of diabetes in the United States during the last three years. Although it is recognized that the arylsulfonylureas have been helpful in the treatment of the mild, stable, maturity-onset type of patient, it was not until the introduction of the biguanide compound, phenformin, that the full range of diabetic classes, from the mild stable adult to the severely labile juvenile, could be successfully treated by oral therapy.' Although the majority of the diabetic population (estimated to be as high as 80% to 85% of all diabetics) are said to fall into the milder adult stable diabetic classification, the physician who treats a large number of diabetics may expect to see a significant number of "problem" diabetic patients, in whom oral therapy had not been effective in the sulfonylurea era. The purpose of this report is to describe a clinical evaluation of phenformin in a group of patients representing almost the entire spectrum of diabetes, emphasizing, however, “the problem case.”

Methods

One hundred and four patients were selected for this series from a private medical practice. Laboratory data obtained on the patients were purposely limited to those common tests which are done by physicians in their own offices. Inasmuch as liver function tests, renal function studies, and hematological studies have been carried out widely on patients treated with phenformin, and the safety of the drug has been adequately documented by many others, these studies were omitted in this survey.

Unlike many prior studies on the clinical effectiveness of oral therapy in diabetes, it will be noted that a large proportion of the present study includes patients heretofore not considered good candidates for oral therapy. For instance, 49% of the patients in this study had insulin requirements in excess of 40 units of insulin per day. Fifty-seven per cent of the patients had onset of diabetes under the age of 40 years. These two groups of patients

Assistant Physician, Chief of the Diabetic Clinic, Hamot Hospital; and Consultant in Diabetes, Erie County Tuberculosis Hospital.

One hundred and four diabetic patients of both sexes, ranging in age from 4 to 83 years, were studied for seven months in a private office practice to evaluate the effect of phenformin therapy. Most of these patients had had difficulty in controlling their diabetes before taking phenformin. Fair to excellent control was achieved by 91 (88%) with or without adjunctive exogenous insulin; failures which occurred in 13 cases were attributable to gastroin testinal, behavioral, or environmental difficulties. Special importance was attached to the value of combined phenformin insulin therapy in reducing the incidence and severity of hypoglycemic episodes and in improving the control of previously labile patients.

have been notoriously poor in their response to the arylsulfonylurea drugs.

Toxicologic studies were purposely omitted so that the primary emphasis was upon the practical clinical use of the drug. All patients in this series were studied in the office without referral to hospital facilities. Hospital admission was not employed to initiate therapy. Urine tests were done, with use of either copper sulfate tablets (Clinitest) or Benedict's method. All tests for blood sugar reported in the series were done by the Folin-Wu micro-method with a photometer. An effort was made to achieve a balance in this study as to sex ratio, duration of diabetes, and age at onset of diabetes and to include in the series groups of patients illustrating the more common problems facing the physician in a private practice emphasizing diabetes management. It will be seen that the sex distribution has been almost equal, with 53 male and 51 female patients.

All patients were carefully evaluated as to degree of control prior to phenformin therapy, by history, physical examination, and laboratory stud

Only a few patients, who were newly diagnosed as diabetic, were started on phenformin therapy when first coming under my care. In the case of patients previously under my care, control had been evaluated through the use of frequent blood sugar determinations, history, and the other criteria usually concerned in the evaluation of control. These criteria are described in greater detail below.

Adults were usually started on a dose of 75 mg. of phenformin per day with only a small reduction (5 to 10%) in insulin dosage. Follow-up visits were scheduled at intervals of one week, two weeks, and finally one month. The patients were furnished with forms for urine-test reporting, and they were requested to test their urine four times daily, before meals and at bedtime, and to keep careful records. Folin-Wu tests for blood sugar were done at the time the patient was seen, at each visit. These values were then correlated with the carbohydrate content of the last meal ingested prior to the office visit, and the time elapsed between the last meal and the test for the blood sugar was carefully noted. Each patient was weighed, and a complete physical examination was performed. The dose of phenformin was usually increased in increments of no higher than 25 mg., until the maximum effective dose was reached. Insulin was reduced as indicated, but no particular pattern was followed in such reduction. Rather, reduction in insulin was gauged by response to phenformin in each case. All patients were instructed to take phenformin at meal times; rarely was a bedtime dose utilized. The most effective method of administering the drug was to start with one 25 mg. tablet three times daily with meals. When increases were necessary, the first increase was with breakfast, the second increase was with the midday meal, and the third increase with the evening meal. For example, those patients taking 125 mg. of phenformin per day took two tablets at breakfast, two with the midday meal, and one with the evening meal. Likewise, those taking 100 mg. per day took two tablets with breakfast, one with lunch, and one at supper

time. It was believed that this pattern of administration most effectively utilized the known therapeutic effects and duration of phenformin. Doses of the drug in this series ranged from a minimum of 37.5 mg. per day to 150.0 mg. per day. Patients who did not respond to 150 mg. of phenformin daily were arbitrarily classified as failures in this group, and no attempt was made to exceed that dosage.

In patients under 15 years of age, the dose of phenformin did not in any instance exceed 100 mg. per day, with a range of 37.5 mg. to 100.0 mg. per day. The average dose in this group was 75 mg. per day. In those patients over 15 years of age, the average dose of phenformin per day was approximately 100 mg., with a range of 50 to 150 mg. per day.

No apology is made for the fact that placebo studies and double-blind techniques were not utilized in the study. In more than one-half of the patients, such techniques would have been fraught with danger. because of the marked lability of the patients. I believed that adequate control studies existed because of my long association with most of the patients before the utilization of phenformin therapy in their management.

Results

Responsiveness to the drug was classified as excellent. good, fair, and failure. Patients classified as having an excellent response had, in all instances, postprandial Folin-Wu tests for blood sugar with results of less than 190 mg.% at all times tested. They were aglycosuric more than 90% of the time. Weight and growth increments in the cases of children remained normal; maintenance of weight and vigor in adult patients was maintained, and there was freedom from side-effects with therapeutically effective dosage of the drug. Patients classified as having a good response had only rare postprandial results of tests for blood sugar over 190 mg.%. Glycosuria was present in less than 20% of tested specimens. Gastrointestinal reactions were ninimal and disappeared as treatment progressed. Weight and growth criteria were met.

Patients classified as having fair response were those in whom the primary benefit derived from phenformin therapy was relief from severe hypoglycemic episodes. These patients showed only general improvement in other areas relating to diabetic control. For instance, blood sugar levels. although not generally under 200 mg. postprandially, were lower than before the use of phenformin. Also there was improvement in weight and growth criteria, in the case of children, and improvement in vigor and sense of well-being in the case of adults. Generally speaking, the patients whose control is listed as fair were those patients in whom control with phenformin therapy was improved but still could not be called satisfactory.

In this group of patients, gastrointestinal sideeffects were also more frequently encountered and more difficult to control, although they did not necessitate the cessation of treatment.

Patients whose response was considered failure in this series fell into two categories. In one category, at the 150-mg. maximum dose level, were those patients who showed lowering effect on the blood sugar level from administration of phenformin. Of the total series of patients, 8% fell into this group. In the second category, an additional 4% of the patients were classified as failures because gastrointestinal side-effects were so severe as to render the continuation of therapy impractical. Therefore, the criteria for failure were lack of hypoglycemic effect from the drug or gastrointestinal side-effects or both. The objective in this study was the evaluation of phenformin as adjuvant therapy. Therefore, only the excellence of control following the use of the oral dose of the drug was considered to be of primary importance. Reduction in insulin requirement was not a factor in

While the replacement of insulin therapy was not a factor in classification of control, and although it was not primarily the effect which I desired in this study, complete replacement of insulin did occur in five patients. Fourteen patients reduced their insulin requirement by as much as 50%. Of 91 patients previously treated with insulin, 73 of the entire series experienced a significant reduction in insulin requirement while taking phenformin. Two patients had only a minor reduction in insulin requirement, 16 had no reduction; and 13 patients had either never received any type of therapy before or had been treated with sulfonylureas.

In the management of diabetes with phenformin therapy, persistence is most important. Although most patients responded to phenformin therapy

within one to two weeks, several patients in this series had no significant response to the drug until four to six weeks after the institution of therapy; in most instances, it was not necessary to increase the level of dosage to obtain response; the patient simply continued with a reasonable dose of phenformin until response became apparent. Of those eight patients who were totally unresponsive to the drug at the 150-mg. maximum dose utilized, therapy was continued for a minimum of three months before failure was acknowledged.

A patient, 14 years of age, with diabetes of one year's duration, showed no appreciable response or improvement in control with use of phenformin until two months after the therapy was instituted. Although, in this instance, reduction of insulin intake was not dramatic (34 units to 30 units daily), control improved remarkably at about the time I was ready to discontinue the drug treatment. Control has continued to be good since that time. This illustrates the virtue of patience when phenformin is used. A too early discontinuance of the drug may result in unwarranted failures.

Selected Case Histories

CASE 1-A 56-year-old man, in whom diabetes was first discovered in 1957, had never received insulin therapy. He had been treated with tolbutamide, in a dose of 1.5 Gm. per day, and diet since the discovery of his diabetes. The patient stated that he had rarely been free from sugar, and over the six months prior to being seen in this office (April 24, 1959), numbness in both feet and severe, aching nocturnal femoral pain had developed. Since the patient's prior diet was considered adequate, he was kept on that diet. The fasting blood sugar level on this date was 198 mg.%. The urine was olive green on Benedict's test. Tolbutamide therapy was discontinued, and the patient was given phenformin, 100 mg. per day. On May 7, one hour after lunch, the blood sugar level was 174 mg.% and the urine was sugar-free. The patient continued to take 100 mg. per day. On May 25, two and one-half hours after lunch, the blood sugar level was 145 mg.% and the urine was sugarfree. On June 15, forty-five minutes after lunch, the capillary blood sugar level was 185 mg.% and the urine was sugar-free.

At this time, the patient stated that he felt better than he had in a year. He had been able to discontinue the use of narcotics for control of his nocturnal leg pains, and the numbness in his feet was considerably improved. The patient stated that he was considering coming out of retirement to take over the active management of his business which he had been forced to relinquish to his sons six months previously because of numbness in his feet, nocturnal pain, and fatigue. On July 13, 45 minutes after lunch, the capillary blood sugar level was again 185 mg.%. The urine was

sugar-free. The patient has had no untoward effects from doses of 100 mg. of phenformin per day, and he continues to have relief from his diabetic neuritis.

CASE 2.-A 20-year-old male patient, whose diabetes began in 1943 at the age of 5 years, had always had poor control. The patient had been hospitalized five times in the previous 15 years for severe diabetic keto-acidosis. Urine tests almost always showed glycosuria. Control was considered poor when the patient was first seen in February, 1959. On March 17, the patient was taking 10 units of regular insulin mixed with 60 units of isophane (NPH) insulin. A two-hour postprandial blood sugar level was 250 mg.%. The patient stated that every urine test done in the prior three weeks showed glycosuria. The patient was given phenformin in doses of 100 mg. per day. The patient has been seen at intervals of two weeks to one month, from March, 1959, through July, 1959, with one-hour postprandial blood sugar levels as follows: 140, 100, 144, 126, and 171 mg.%. The patient has been aglycosuric more than 90% of the time. Weight has been maintained. There have been no episodes of hypoglycemia. The patient has had no side-effects on therapy with phenformin in a dosage of 125 mg. per day. Control has been considered outstanding. The insulin dosage has dropped to 30 units of isophane insulin daily.

CASE 3.-A 64-year-old man, a railroad engineer, noted the gradual onset of thirst, excessive urination, fatigue, and nocturia approximately six months prior to being seen in this office on May 15, 1959. At that time, the patient's capillary blood sugar level, three and one-half hours after his noonday meal, was 251 mg.%. The urine was brick red on Benedict's test. Physical examination revealed a blood pressure of 144/90 mm. Hg. He weighed 2271⁄4 lb. (103.27 kg.). A fistula in ano was noted. The patient was started on phenformin therapy, in doses of 150 mg. per day and a diabetic weight-reduction diet. On May 26, the patient was seen one hour after breakfast. His blood sugar level was 115 mg.%. The urine was sugarfree. On June 16, two hours after lunch, the blood sugar level was 145 mg.%. The urine was sugarfree. On July 6, the patient underwent a fistulectomy in the hospital. No insulin was used in the preoperative or postoperative period. Postprandial blood sugar levels did not exceed 160 mg.%. On July 16, one hour after a meal, the blood sugar level was 167 mg.%, and the urine was again sugarfree. On his weight-reduction diet, the patient's weight dropped from 2271⁄4 lb. to 2031⁄4 lb. (92.3 kg.). At the time of writing the patient feels well, and has maintained his vigor, and has been able to continue his usual duties. This patient has achieved excellent control on phenformin therapy alone and has been able to go through surgery without the use of insulin. Excellent control on phenformin

and diet continued when last seen on Sept. 8, 1959. CASE 4.-A 92-year-old boy had diabetes which started in January, 1955, at the age of 5 years and 9 months. In spite of excellent cooperation on the part of his parents and frequent insulin adjustments, the patient's diabetes had been extremely poorly controlled. The patient had had frequent severe insulin reactions. When first seen in January, 1959, at the age of 91⁄2 years, the patient weighed 59 lb. (27 kg.) and his height was 524 in. (122.7 cm.). The patient was taking 9 units of protamine zinc insulin mixed with 19 units of regular insulin. Since being given phenformin, in a dosage of 75 mg. per day, with 4 units of regular insulin mixed with 16 units of isophane insulin, the patient's control has been much improved. Three-hour postprandial blood sugar levels have been as follows: 76, 203, 60, 56, and 216 mg.%. In spite of the low blood sugar levels, the patient has had no hypoglycemic symptoms. In the seven months that the patient has been on combined phenformin and insulin therapy, he has grown 11⁄2 in. (3 cm.) and has gained 4 lb. (1.8 kg.), whereas his growth in the previous year, on insulin therapy alone, had been 1⁄2 in. without weight gain.

CASE 5.-A 142-year-old boy had diabetes which began in 1950, at the age of five. The patient's control always had been poor, with widely fluctuating blood sugar levels and frequent convulsivetype episodes of hypoglycemia. The patient had been in severe diabetic keta-acidosis in 1951. The patient is a psychopathic individual who has constantly battled with authority. He has a poor school record. This patient exemplifies the worst possible adjustment to the impact of diabetes in the juvenile individual. He was given phenformin in doses ranging from 50 to 100 mg. per day. There were no gastrointestinal side-effects. During the six months that the patient was on phenformin and insulin therapy, no improvement in the patient's control was noted. In this case, although rated as a failure of phenformin therapy, the effect of the patient's lack of cooperation must be taken into

account.

CASE 6.-A 51-year-old woman had diabetes beginning in 1929, at the age of 21. The patient is a cooperative and intelligent school teacher who has managed her diabetes well. The patient has had frequent severe insulin reactions coming on without warning during the five years prior to the time of this writing. She has noted cyclical variations in insulin requirement, ranging from 30 to 50 units per day. When first given phenformin in March, 1959, the patient was taking 24 units of isophane insulin daily. Good hypoglycemic response was noted with doses of 100 mg. of phenformin per day, but the patient had anorexia, occasional episodes of nausea and vomiting, and constant gastrointestinal distress even when the dosage was re