Clinical outcomes and predictors of success with Impella weaning in cardiogenic shock: a single-center experience.

Introduction: Cardiogenic shock (CS) is a severe syndrome with poor prognosis. Short-term mechanical circulatory support with Impella devices has emerged as an increasingly therapeutic option, unloading the failing left ventricle (LV) and improving hemodynamic status of affected patients. Impella devices should be used for the shortest time necessary to allow LV recovery because of time-dependent device-related adverse events. The weaning from Impella, however, is mostly performed in the absence of established guidelines, mainly based on the experience of the individual centres. Methods: The aim of this single center study was to retrospectively evaluate whether a multiparametrical assessment before and during Impella weaning could predict successful weaning. The primary study outcome was death occurring during Impella weaning and secondary endpoints included assessment of in-hospital outcomes. Results: Of a total of 45 patients (median age, 60 [51-66] years, 73% male) treated with an Impella device, 37 patients underwent impella weaning/removal and 9 patients (20%) died after the weaning. Non-survivors patients after impella weaning more commonly had a previous history of known heart failure (p = 0.054) and an implanted ICD-CRT (p = 0.01), and were more frequently treated with continuous renal replacement therapy (p = 0.02). In univariable logistic regression analysis, lactates variation (%) during the first 12-24 h of weaning, lactate value after 24 h of weaning, left ventricular ejection fraction (LVEF) at the beginning of weaning, and inotropic score after 24 h from weaning beginning were associated with death. Stepwise multivariable logistic regression identified LVEF at the beginning of weaning and lactates variation (%) in the first 12-24 h from weaning beginning as the most accurate predictors of death after weaning. The ROC analysis indicated 80% accuracy (95% confidence interval = 64%-96%) using the two variables in combination to predict death after weaning from Impella. Conclusions: This single-center experience on Impella weaning in CS showed that two easily accessible parameters as LVEF at the beginning of weaning and lactates variation (%) in the first 12-24 h from weaning begin were the most accurate predictors of death after weaning.

Vitamin E intake reduces plasminogen activator inhibitor type 1 in T2DM patients.

Previous studies hypothesised that vitamin E could protect against coronary heart disease and vascular complications in diabetes, but no studies have been performed regarding its eventual effects on fibrinolysis. Nevertheless, in Type 2 diabetes mellitus (T2DM) a profound reduction in the fibrinolytic activity has been demonstrated to be involved in vascular complications, probably due to plasminogen activator inhibitor type 1 (PAI-1) overproduction. On this basis we aimed to verify whether an antioxidant treatment with vitamin E is able to lower PAI-1 plasma levels in T2DM. Thirteen T2DM patients (9 males and 4 females; mean age+/-SD, 64.4+/-3.3 yr) were selected through strict admission criteria. These patients were treated with vitamin E (500 IU/die) for 10 weeks. Glyco-lipometabolic, oxidative and haemocoagulative parameters were evaluated at baseline and after 5, 10, 30 and 60 weeks. Vitamin E levels at different times were [median (interquartile range)] 6.1 (5.3-7.7), 8.5 (7.3-9.9), 9.7 (8.9-12.9), 5.6 (4.4-6.8), 5.7 (4.5-7.1) microg/ml, respectively. Significant differences were found for PAI-1 antigen (p=0.006), PAI-1 activity (p=0.028), apolipoprotein B (p=0.015) and antioxidant defence, evaluated as ferric reducing ability of plasma (FRAP) values (p=0.005). Particularly, decrements were detected for PAI-1 antigen between baseline and the 10th week (p<0.05), followed by an increase back to basal at the 30th week. Similar behaviour was found for PAI-1 activity. Regarding the antioxidant defence, FRAP values increased until the 30th week (p<0.05) with a decrease at the 60th week. These results demonstrate that vitamin E is able to lower PAI-1 levels in diabetic patients but this effect does not seem related to improvements of glycometabolic data or to the increase in FRAP values, suggesting that PAI-1 overproduction can be decreased by other effects of vitamin E on endothelial cells.

Fibronectin and lipoprotein(a) are inversely related to plasminogen activator inhibitor type-1 levels in Type 2 diabetic patients without complications.

Plasminogen activator inhibitor type-1 (PAI-1), the most important physiological fibrinolysis inhibitor, is considered an independent factor of cardiovascular risk in Type 2 diabetes mellitus (T2DM). In previous papers we demonstrated that a T2DM population without complications presents: 1) PAI-1 not increased with respect to a control group; and 2) a negative correlation between PAI-1 and lipoprotein(a) [Lp(a)], suggesting that in these subjects PAI-1 levels could be modulated by the "endothelial stress" induced by Lp(a) and diabetes. This work has been performed in order to better verify this intriguing hypothesis, and the endothelial stress has been evaluated through a marker of endothelial damage, fibronectin (FNC). For this purpose we chose a T2DM population without complications (n=73) and a control group (n=46). Plasma concentrations of FNC, Lp(a), PAI-1 antigen and activity, and the main parameters of lipo- and glycometabolic balance were determined. Fibronectin was significantly higher in diabetics with respect to controls (p<0.01). As expected, significant correlation between PAI-1 antigen, PAI-1 activity and Lp(a) (r=-0.54,p<0.01 and r=-0.39,p<0.01, respectively) was found only in diabetic patients. In the same group FNC showed a significant correlation with PAI-1 antigen and activity (r=-0.49,p<0.01 and r=-0.47; p<0.01, respectively), while no relationship was found between Lp(a) and FNC. Multiple regression analysis showed statistically significant correlation between PAI-1 antigen and PAI-1 activity with FNC and Lp(a) in diabetic patients without complications (p<0.05). These data suggest that in absence of complications, the endothelium is able to modulate PAI-1 levels, favouring in that way the fibrinolytic pathway and, subsequently, the recovery of the endothelial integrity. This modulation seems to be related to parameters such as Lp(a) and FNC, although the mechanisms of the endothelial stress of these two molecules seem to be different.

Long-term intensive treatment of type 1 diabetes with the short-acting insulin analog lispro in variable combination with NPH insulin at mealtime.

OBJECTIVE: To establish whether the short-acting insulin analog lispro can be successfully implemented in long-term intensive insulin therapy in type 1 diabetes, and if so, what its effects are on glycemic control and frequency and awareness of hypoglycemia. RESEARCH DESIGN AND METHODS: We randomized 56 type 1 diabetic patients to treatment with either lispro (n = 28) or human regular insulin (Hum-R; n = 28) as mealtime insulin for 1 year (open design, parallel groups). Lispro was injected at mealtime and Hum-R was given 10-40 min before meals (bedtime NPH was continued on both occasions). With lispro, NPH was added at breakfast (approximately 70/30), lunch (approximately 60/40), and supper (approximately 80/20) (mixing percentage of lispro/NPH) to optimize premeal and bedtime blood glucose. RESULTS: Total daily insulin units were no different in the two treatment groups, but with lispro approximately 30% less short-acting insulin at meals and approximately 30% more NPH was needed versus Hum-R (P < 0.05). The bedtime NPH dosage was no different. With lispro + NPH, the mean daily blood glucose was lower than with Hum-R (8.0 +/- 0.1 vs. 8.8 +/- 0.1 mmol/l; P < 0.05), HbA1c was lower (6.34 +/- 0.10 vs. 6.71 +/- 0.11%, mean value over 1 year; P < 0.002), and hypoglycemia (blood glucose < or = 3.8 mmol/l) was less frequent (7.4 +/- 0.5 vs. 11.5 +/- 0.7 episodes/patient-month) and tended to occur more within 90 min after meals than in the postabsorptive state (P < 0.05 vs. Hum-R). After 1 year, plasma adrenaline and symptom responses to experimental, stepped hypoglycemia improved with lispro and were closer to the responses of 12 nondiabetic control subjects versus Hum-R both in terms of thresholds and magnitude (P < 0.05). CONCLUSIONS: We concluded that mealtime injection of lispro + NPH improves the 24-h blood glucose and the percentage HbA1c as compared with Hum-R. The improvement can be maintained long term. Intensive therapy with lispro + NPH results in less frequent hypoglycemia and better awareness and counterregulation of hypoglycemia.

A strong inverse relationship between PAI-1 and Lp(a) in hypertensive Type 2 diabetic patients.

Thrombophilia with a contemporary reduction of fibrinolytic activity has been observed both in diabetes mellitus and hypertension. Previously, we found a relationship between plasminogen activator inhibitor Type 1 (PAI-1) and lipoprotein(a) [Lp(a)] in Type 2 diabetes mellitus patients without complications. We hypothesised that this relationship could be due to a compensatory mechanism able to lower the risk of hypofibrinolysis as found in Type 2 diabetes mellitus. The present work was aimed at investigating the influence of concurrent hypertension and diabetes mellitus on the plasma levels of these two fibrinolytic inhibitors. In addition, other risk factors, known to influence the fibrinolytic parameters, were taken into account. Forty-nine Type 2 nonhypertensive diabetic patients without complications, 47 Type 2 hypertensive diabetic patients without complications, 54 non-diabetic hypertensive subjects without complications as well as 87 control subjects were studied. Plasma concentrations of Lp(a), PAI-1 antigen and activity, and the main parameters of oxidative, lipo- and glycometabolic balance were determined. Significant statistical differences between diabetic and non-diabetic subjects were found concerning triglycerides and antioxidant defence (p<0.01). Analysis of variance showed the F test statistically significant in evaluating the Log PAI-1/Lp(a) (p = 0.02). Correlation analysis between Log PAI-1 antigen and Lp(a) was significant in non-hypertensive diabetic patients, as expected (r = -0.38, p<0.01), and even stronger in hypertensive diabetic patients (r = -O.72,p<0.01). These results allow to hypothesise that the relationship between PAI-1/Lp(a) could be determinant in avoiding vascular complications due to diabetes mellitus and hypertension.

In vitro activity of terbinafine against clinical isolates of dermatophytes.

A broth macrodilution method following the recommendations established by the National Committee for Clinical Laboratory Standards was employed for testing terbinafine against 20 clinical isolates of dermatophytes belonging to seven species. The minimal inhibitory concentrations resulting in either 80% or 100% inhibition of growth, compared to growth in drug-free control tubes, ranged from

Comparison of three methods for testing azole susceptibilities of Candida albicans strains isolated sequentially from oral cavities of AIDS patients.

Three susceptibility testing procedures were compared to determine fluconazole, itraconazole, and ketoconazole MICs against 47 Candida albicans strains isolated sequentially from the oral cavities of five AIDS patients undergoing azole therapy. They included the broth microdilution method (BM), performed according to the National Committee for Clinical Laboratory Standards' tentative standard, the agar dilution method (AD), and the Etest; the latter two tests were performed both in Casitone agar (AD-Cas and Etest-Cas) and in RPMI (AD-RPMI and Etest-RPMI). Twenty-four- and 48-h MICs obtained by AD and Etest were compared with 48-h MICs obtained by BM. The MICs of all the azoles determined by BM were usually lower than those obtained by the other methods, mainly due to different reading criteria. In order to assess the most appropriate way of evaluating the agreement of MICs obtained by different methods with those produced by the proposed reference method (BM), we used the mean differences calculated according to Bland and Altman's method. Comparison of fluconazole MICs obtained by BM and AD-Cas yielded a mean difference of 3, and the percentages of agreement within +/-2 dilutions were 98 and 100% at 24 and 48 h, respectively. For ketoconazole and itraconazole MICs, lower mean differences were noted, and agreement ranged from 96 to 100%. Agreement between the AD-RPMI and BM results was poor for all azoles, and an increase in MICs was always observed between the 1st- and 2nd-day readings. Similarly, Etest-Cas gave better agreement with BM than did Etest-RPMI for all the azoles. BM, AD-Cas, and Etest-Cas each demonstrated a progressive increase in fluconazole MICs against strains isolated sequentially from a given patient, in accordance with the decreased clinical response to fluconazole.

In-vitro interaction of terbinafine with amphotericin B, fluconazole and itraconazole against clinical isolates of Candida albicans.

A chequerboard titration broth microdilution method, performed according to the recommendations of the National Committee for Clinical Laboratory Standards, was applied to study the in-vitro interaction of terbinafine with amphotericin B, fluconazole and itraconazole against 30 strains of Candida albicans isolated from the oral cavities of AIDS patients. MICs were determined spectrophotometrically at 490 nm and read at either 24 h or 48 h. The end-point was defined as the drug concentration resulting in > or = 90% inhibition of growth relative to control growth. Synergy, defined as a fractional inhibitory concentration (FIC) index of < or = 0.50, was observed in 93% (28 of 30) of terbinafine-amphotericin B interactions, in 47% (14 of 30) of terbinafine-fluconazole interactions and in 43% (13 of 30) of terbinafine-itraconazole interactions; antagonism (FIC > 2.0) was not observed. Where synergy was not achieved, there was still a decrease, although not as dramatic, in the MIC of one or both drugs when used in combination. Reading the MICs on day 2 did not significantly affect the mode of interaction of terbinafine-triazoles, while for terbinafine-amphotericin B the proportion of synergic interactions dropped from 93% (28 of 30) to 30% (nine of 30; P = 0.0001). Antagonism was not observed for any drug combination even at 48 h. Minimum fungicidal concentrations (MFCs) of all drugs alone and in combination were determined against five isolates. Neither terbinafine nor the two triazoles showed fungicidal activity when tested alone or in combination. The fungicidal activity of amphotericin B was slightly enhanced when combined with terbinafine, there being a decrease of two-fold dilutions in the amphotericin B MFCs against all five isolates tested. Thus terbinafine enhances the activities of amphotericin B and triazoles against C. albicans in vitro. Clearly, clinical studies are warranted to elucidate further the potential utility of these combination therapies.

In vitro activity of five antifungal agents against clinical isolates of Saccharomyces cerevisiae.

We evaluated the in vitro activity of fluconazole, itraconazole, ketoconazole, 5-fluorocytosine and amphotericin B against 30 clinical isolates of Saccharomyces cerevisiae by a broth microdilution method, following the NCCLS recommendation. Testing was performed either in RPMI-1640 or yeast nitrogen base (YNB). YNB supported the growth of all isolates tested, while results in RPMI-1640 were not obtained for six isolates (20%). The MIC of all three azoles in YNB were one or two dilutions higher than those obtained in RPMI-1640 (P=0.0001 for fluconazole and itraconazole, P=0.03 for ketoconazole). Elevated MICs were observed for all three azoles, while all the isolates were susceptible to 5-fluorocytosine and amphotericin B. All MIC values were confirmed by spectrophotometric reading. Six strains of S. cerevisiae isolated from the faeces and consecutive blood cultures from an AIDS patient over a 7-month period were typed by electrophoretic karyotyping (EK). EK showed the maintenance of the same karyotype over time suggesting that the faecal isolate changed from a colonizing to infection-causing strain. The relative resistance of S. cerevisiae to azole drugs as well as its ability to cause widespread infections may promote the emergence of this species as a pathogen in immunosuppressed patients.

Genotypic identification of sequential Candida albicans isolates from AIDS patients by polymerase chain reaction techniques.

Random amplification of polymorphic DNA and inter-repeat polymerase chain reaction (IR-PCR) were compared with restriction fragment length polymorphism (RFLP) analysis as methods for DNA typing of Candida albicans. Forty-seven strains of Candida albicans isolated from the oral cavities of five AIDS patients undergoing fluconazole therapy were analyzed. There was an excellent correspondence between the DNA types obtained by both PCR-based techniques and by RFLP. With the exception of one patient who was infected with three DNA types of Candida albicans during a five-year observation period, the patients each harboured only one major strain, which became progressively less susceptible to fluconazole. Each DNA type was unique to a patient. The data suggest that these typing methods are suitable for investigating the epidemiology of oropharyngeal candidiasis in this patient group.

Effects of recent, short-term hyperglycemia on responses to hypoglycemia in humans. Relevance to the pathogenesis of hypoglycemia unawareness and hyperglycemia-induced insulin resistance.

A single episode of recent hypoglycemia increases, whereas long-term hyperglycemia decreases, the glycemic thresholds of responses of counterregulatory hormone and symptoms to subsequent hypoglycemia in humans. To assess whether short-term, antecedent hyperglycemia exerts effects opposite to those observed after acute hypoglycemia, seven normal, nondiabetic subjects and eight insulin-dependent diabetes mellitus (IDDM) patients were studied during hyperinsulinemic-hypoglycemic clamp (sequential, 90-min plateaus of plasma glucose [PG] of 4.3, 3.7, 3.0, and 2.4 mmol/l). Nondiabetic subjects were studied the morning after either 6-h clamped hyperglycemia (PG approximately 13.5 mmol/l) or euglycemia (PG approximately 5 mmol/l) between 1600 and 2200 the previous day (glucose and insulin infused on both occasions), as well as after nocturnal hyperglycemia (PG approximately 13.5 mmol/l) or euglycemia between 2300 and 0500. The IDDM patients were studied after 15 h of euglycemia or hyperglycemia (approximately 17 mmol/l) but identical hyperinsulinemia (approximately 225 pmol/l) between 1600 and 0700. Neither PG thresholds of counterregulatory hormone, symptoms, onset of cognitive dysfunction to hypoglycemia, nor maximal responses were affected by antecedent, short-term hyperglycemia in normal nondiabetic subjects and IDDM patients (NS). However, the rate of glucose infusion required to maintain hypoglycemic plateaus during hypoglycemia was lower after hyperglycemia (nondiabetic subjects 31.2 +/- 3.4 vs. 36.7 +/- 4 mumol.kg-1.min-1, IDDM patients 33 +/- 3.1 vs. 42.5 +/- 3.9 mumol.kg-1.min-1; P < 0.05) indicating greater insulin resistance induced by antecedent hyperglycemia. In conclusion, in contrast to acute hypoglycemia and long-term hyperglycemia, recent, short-term hyperglycemia does not affect physiological responses to hypoglycemia.(ABSTRACT TRUNCATED AT 250 WORDS)

A risk-benefit appraisal of acarbose in the management of non-insulin-dependent diabetes mellitus.

Acarbose is an alpha-glucosidase inhibitor proposed for the treatment of diabetic patients. It acts by competitively inhibiting the alpha-glucosidases in the intestinal brush border. The principal action of these enzymes is to convert nonabsorbable dietary starch and sucrose into absorbable monosaccharides (e.g. glucose). Enzyme inhibitors delay this conversion, slowing the formation and consequently the absorption of monosaccharides, and thus reducing the concentration of postprandial blood glucose. Both starch and sucrose are influenced, whereas lactose and glucose are not. Many studies in experimental animals, healthy volunteers and patients with non-insulin-dependent diabetes mellitus (NIDDM) have shown that acarbose decreases postprandial blood glucose, with a lesser reduction of fasting blood glucose, plasma triglycerides and postprandial insulin levels. In long term studies in NIDDM patients, acarbose significantly reduced glycosylated haemoglobin levels. Acarbose is only minimally absorbed from the gut and no systemic adverse effects have been demonstrated after long term administration. The drug allows undigested carbohydrates to pass into the large bowel where they are fermented causing flatulence, bloating and diarrhoea. These symptoms, which occur in approximately 30 to 60% of patients, tend to decrease with time and seem to be dose-dependent. They are minimised by starting therapy with low doses (such as 50mg 3 times daily) which may be effective in many patients. An increase in serum hepatic transaminases observed in earlier studies in the US, where doses of acarbose up to 900mg daily were used, has been not reported with the lower doses of the drug actually recommended [150 to 300mg (up to 600mg) daily]. In conclusion, acarbose may be useful in patients with NIDDM when diet alone is no longer able to maintain satisfactory blood glucose control. Furthermore, it may be a valid alternative to sulphonylurea or biguanide therapy when these drugs are contraindicated and insulin administration may be delayed. Acarbose seems also to be a useful adjunct to hypoglycaemic oral agents but its precise role in this field has not been fully clarified.

[Oral glucose tolerance and HbA1c in the diagnosis of pregnancy in diabetes]. / Carico orale di glucosio ed HbA1c nella diagnosi di diabete gestazionle.

This study was undertaken in order to evaluate the plasma glucose response to oral glucose tolerance test (OGTT, 100 g) and the HbA1c values in pregnant women at different gestational ages. One-hundred twenty-nine OGTTs have been performed in 75 pregnancies. The results obtained show a decrease in glucose tolerance during pregnancy. Mean HbA1c value was significantly higher in women with gestational diabetes mellitus, but values of subjects with gestational diabetes and normal glucose tolerance overlapped widely. In conclusion, HbA1c is not a sensitive parameter in the diagnosis of gestational diabetes mellitus. Further studies are necessary to evaluate its specificity and prognostic significance.

[Importance of CA-50 detection in malignant gynecologic tumors]. / Importanza della determinazione del CA-50 nei tumori ginecologici maligni.

The role of Tumor Associated Antigens in the diagnosis of neoplasia is now actively investigated. The CA-50 antigen is known to be elevated in different types of gynecological neoplasia. In this study the CA-50 levels have been measured in 70 patients with ovarian, endometrial or cervical cancer. The data obtained show that the CA-50 levels, although elevated above normal in a high percentage of the patients studied, cannot be considered diagnostic for the presence of neoplasia.

[Corticosteroids]. / Corticosteroidi.

Adrenal steroids, particularly glucocorticoids, are used in a variety of conditions ranging from adrenal insufficiency requiring substitution therapy (glucocorticoids and mineral-corticoids), to a wide range of clinical disorders in which undesired inflammatory reactions must be reduced (glucocorticoids). The anti-inflammatory effect, typical of glucocorticoids, is mainly due to the interference with arachidonic acid metabolism, from which both prostaglandins and leukotriens, mediators of inflammation, take origin. ACTH is also employed with the same indications of glucocorticoids, but its use is limited, since its effect depends on the stimulation of cortico-adrenal gland with subsequent release of not only glucocorticoids but also mineral-corticoids and androgens. The therapeutical indications of glucocorticoids are numerous, but dose and duration of treatment vary in relation of the disease that is to be cured. The widespread use of steroids is accompanied by numerous unwanted reactions which depend on their metabolic actions. Therefore corticosteroids must be reserved to well defined clinical conditions and their use should be guided by criteria that are now codified. In this review the physiological and pharmacological effects and the clinical use of corticosteroids have been examined.

Improved column-chromatographic isolation of glycosylated hemoglobin S.

We investigated glycosylated hemoglobin S by means of chromatography on Bio-Rex 70. The selected elution conditions were similar to those described by Trivelli et al. (N Engl J Med 284: 353-357, 1971), except for modified ionic strength and accurate temperature control. This enabled us to isolate a minor hemoglobin fraction whose properties, as determined by chromatography, electrophoresis, and two-dimensional maps of its tryptic peptides, were typical of a hemoglobin S tetramer with blocked N-terminal residues of the beta subunits. The colorimetric test indicated this to be a glycosylated hemoglobin. This procedure is notably improved over previous chromatographic techniques for isolating glycosylated hemoglobin S. Moreover, it can be easily used in any laboratories where Trivelli's method is routinely in use.

Mechanisms of glucagon secretion during insulin-induced hypoglycemia in man. Role of the beta cell and arterial hyperinsulinemia.

To elucidate the mechanisms controlling the response of glucagon to hypoglycemia, a vital component of the counterregulatory hormonal response, the role of intraislet insulin was studied in seven normal subjects and five subjects with insulin-dependent diabetes mellitus (IDDM) (of less than 15-mo duration). In the normal subjects, hypoglycemia (arterial plasma glucose [PG] 53 +/- 3 mg/dl) induced by an intravenous insulin infusion (30 mU/m2 X min for 1 h, free immunoreactive insulin [FIRI] 58 +/- 2 microU/ml) elicited a 100% fall in insulin secretion and an integrated rise in glucagon of 7.5 ng/ml per 120 min. When endogenous insulin secretion was suppressed by congruent to 50 or congruent to 85% by a hyperinsulinemic-euglycemic clamp (FIRI 63 +/- 1.5 or 147 +/- 0.3 microU/ml, respectively) before hypoglycemia, the alpha cell responses to hypoglycemia were identical to those of the control study. When the endogenous insulin secretion was stimulated by congruent to 100% (hyperinsulinemic-hyperglycemic clamp, FIRI 145 +/- 1.5 microU/ml, PG 132 +/- 2 mg/dl) before hypoglycemia, the alpha cell responses to the hypoglycemia were also superimposable on those of the control study. Finally, in C-peptide negative diabetic subjects made euglycemic by a continuous overnight intravenous insulin infusion, the alpha cell responses to hypoglycemia were comparable to those of normal subjects despite absent beta cell secretion, and were not affected by antecedent hyperinsulinemia (hyperinsulinemic-euglycemic clamp for 2 h, FIRI 61 +/- 2 microU/ml). These results indicate that the glucagon response to insulin-induced hypoglycemia is independent of the level of both endogenous intraislet and exogenous arterial insulin concentration in normal man, and that this response may be normal in the absence of endogenous insulin secretion, in contrast to earlier reports. Thus, loss of beta cell function is not responsible for alpha cell failure during insulin-induced hypoglycemia in IDDM.