Normal Therapeutic Range Research Articles

Zolpidem and Driving Impairment

Zolpidem, a non-benzodiazepine hypnotic, was identified in the blood of 29 subjects arrested for impaired driving. Zolpidem concentrations ranged from 0.05 to 1.4 mg/L (mean 0.29 mg/L, median 0.19 mg/L). In the subjects whose cases we reviewed where zolpidem was present with other drugs and/or alcohol, symptoms reported were generally those of CNS depression. Symptoms included slow movements and reactions, slow and slurred speech, poor coordination, lack of balance, flaccid muscle tone, and horizontal and vertical gaze nystagmus. In five separate cases, where zolpidem was the only drug detected (0.08-1.40 mg/L, mean 0.65 mg/L, median 0.47 mg/L), signs of impairment included slow and slurred speech, slow reflexes, disorientation, lack of balance and coordination, and "blacking out." Although no quantitative relationship between blood concentrations and degree of driving impairment is currently possible, it is reasonable to conclude that because of its specific activity as a sleep inducer, blood concentrations consistent with therapeutic doses of zolpidem have the potential to affect driving in a negative way, and that concentrations above the normal therapeutic range would further impair a person's level of consciousness and driving ability.

Is there a pharmacist in the house?

Is there a pharmacist in the house?

Carisoprodol, Meprobamate, and Driving Impairment

This paper considers the pharmacology of the centrally acting muscle relaxant carisoprodol, and its metabolite meprobamate, which is also administered as an anxiolytic in its own right. Literature implicating these drugs in impaired driving is also reviewed. A series of 104 incidents in which these drugs were detected in the blood of drivers involved in accidents or arrested for impaired driving was considered, with respect to the analytical toxicology results, patterns of drug use in these subjects, the driving behaviors exhibited, and the symptoms observed in the drivers. Symptomatology and driving impairment were consistent with other CNS depressants, most notably alcohol. Reported driving behaviors included erratic lane travel, weaving, driving slowly, swerving, stopping in traffic, and hitting parked cars and other stationary objects. Drivers on contact by the police displayed poor balance and coordination, horizontal gaze nystagmus, bloodshot eyes, unsteadiness, slurred speech, slow responses, tendency to doze off or fall asleep, difficulty standing, walking or exiting their vehicles, and disorientation. Many of these cases had alcohol or other centrally acting drugs present also, making difficult the attribution of the documented impairment specifically to carisoprodol and meprobamate. In 21 cases, however, no other drugs were detected, and similar symptoms were present. Impairment appeared to be possible at any concentration of these two drugs; however, the most severe driving impairment and most overt symptoms of intoxication were noted when the combined concentration exceeded 10 mg/L, a level still within the normal therapeutic range.

キャピラリー電気泳動法によりフェニトイン中毒を同定した一症例

A 2-year and 5-month old male child was admitted to our hospital, deu to signs of antiepileptic drug intoxication. He had been administered zonisamide and valproic acid for the treatment of epilepsy. One day after taking the medicines prepared in an outside pharmacy according to the doctor's prescription, he showed ataxia. The serum valproic acid concentration was within the normal therapeutic range, when measured using the fluorescence polarization immunoassay method (FPIA). We assayed the zonisamide concentration in patient serum using the capillary electrophoresis method. The peak of zonisamide was very low, but another extremely high peak was found. This peak was identified as phenytoin based on the migration time and absorption spectrum. We therefore concluded that the patient fell into phenytoin intoxication by taking a large amount of phenytoin that had been incorrectly prepared. The capillary electrophoresis method appears to be useful for therapeutic drug monitoring, especially in drug intoxication, because of its specificity of separation, simultaneous determination, speed of analysis, and small injection volume.

Homocysteine as a risk factor for atherosclerosis.

To review the role of homocysteine as a risk factor in the pathogenesis of atherosclerosis and to provide recommendations for the treatment of hyperhomocysteinemia. A MEDLINE search using key terms such as homocysteine, atherosclerosis, folic acid, vitamin B6, and vitamin B12 was conducted for the time period 1966 through January 1999. An article was selected for inclusion in this review if it assessed the relationship and proposed mechanisms of hyperhomocysteinemia on the vasculature, physiologic changes due to hyperhomocysteinemia, and outcomes due to hyperhomocysteinemia, such as morbidity and mortality. In addition, studies that assessed the treatment outcomes of hyperhomocysteinemia were evaluated. Studies of patients with cerebral vascular disease reveal elevated homocysteine concentrations in 30-40% of patients compared with controls. Many studies demonstrate a correlation between elevated homocysteine concentrations, risk of myocardial infarction, and mortality. In addition, hyperhomocysteinemia and decreased folic acid concentrations have been identified in end-stage renal disease (ESRD) and type 2 diabetic patients, while both concentrations remained normal in healthy controls. Studies using folic acid 650 microg/d reduced homocysteine concentrations to within normal therapeutic range after two weeks of treatment. Studies with vitamins B6 and B12 have demonstrated that the use of either alone is ineffective, but when combined or administered with folic acid, homocysteine concentrations return to normal. All therapies must be given for the lifetime of the patient. In addition, patients must use discretion in their diet, as common beverages, such as coffee, have a strong correlation with hyperhomocysteinemia, while foods high in folic acid, vitamin B6 and vitamin B12 may reduce homocysteine concentrations. Additional prospective studies are needed to determine effects of treatment of hyperhomocysteinemia and various diets on atherosclerotic morbidity and mortality. Studies demonstrate a positive correlation between hyperhomocysteinemia and atherosclerosis. The treatment of choice for hyperhomocysteinemia is folic acid. Although the optimal dose is not known, 650 microg/d is the minimum effective dose. To date, no studies have assessed the effects on morbidity and mortality when treating high homocysteine concentrations in atherosclerotic patients.

服薬指導における症例報告 その２ ピルメノールにより肝障害を引き起こしたと疑われた一症例およびその検討

A 56-year-old woman inpatient that had been administrated warfarin, digoxin, verapamil and disopyramide after undergoing surgery for a mitral and aortic valve replacement associated with artrial fibrillation received a 300mg daily dose of disopyramide therapy before and after the operation. Although the serum disopyramide concentration was within the normal therapeutic range, dry mouth appeared as a side effect. Disopyramide was thus changed to pirmenol. The trough level of pirmenol was 2.1μg/mL at seven days after starting pirmenol therapy at the dose of 300 mg/day, and the dose was there after decreased to 200mg/day. About two weeks after pirmenol therapy was started, liver injury was observed. At approximately 30 days after pirmenol administration was stopped, the liver function returned to a normal level.On the other hand, according to recent reports pirmenol was suggested to show a high level in the liver. The reported levels of pirmenol were also similar to for amiodarone and aprindine, which are both well known to induce side effect in the liver. Therefore, one of the reasons that pimenol induced liver injury may be due to its high levels in the liver.

INDUCTION OF PROSTATE APOPTOSIS BY DOXAZOSIN IN BENIGN PROSTATIC HYPERPLASIA

Purpose The molecular mechanisms underlying the therapeutic effect of the alpha 1 blocker, doxazosin, on benign prostatic hyperplasia (BPH) are poorly understood. We evaluated the effect of doxazosin on cell proliferation and apoptosis in the prostatic glandular epithelium and stroma of patients with BPH. Materials and Methods We examined proliferative and apoptotic activities in prostate specimens of 22 men a mean of 65 years old with BPH before and after doxazosin treatment within the normal therapeutic range. Proliferative and apoptotic indexes were determined using Ki-67 nuclear antigen staining and the terminal transferase end labeling assay, respectively. The smooth muscle cell content in prostatic specimens was identified by smooth muscle alpha-actin, and desmin immunoreactivity and apoptotic indexes were correlated with prostatic stromal tissue regression and improvement in BPH symptoms. Results In response to doxazosin treatment there were no significant changes in the kinetics of cell proliferation in the prostatic epithelial or stromal cell population. Mean pretreatment baseline apoptosis was 1.9 and 1.0% for the epithelial and stromal prostate components, respectively. Mean apoptotic indexes significantly increased after 3 months of doxazosin treatment in the glandular epithelial (6%) and smooth muscle cells (15%). By 12 months after treatment epithelial apoptosis had decreased to constitutive levels, while the apoptotic index of prostatic stroma cells remained high. Doxazosin induced smooth muscle cell apoptosis correlated with prostatic stromal degeneration, decreased alpha-smooth muscle actin expression and improved BPH symptoms. Conclusions These findings implicate the induction of prostate apoptosis by doxazosin as a potential molecular mechanism underlying the acute and chronic therapeutic responses of BPH to alpha 1 blockade.

ジゴキシンの有効治療域内で視覚異常が出現した１症例

A 65-year-old woman inpatient had been administered methyldigoxin, verapamil and warfarin for the treatment of artrial fibrillation and mitral stenosis disease. About six day after the start of treatment the dose of methyldigoxin was increased from 0.1 to 0.2 mg/day and, as a result, a visual disturbance occurred. However, no other side effects of digoxin were not presented and the serum digoxin concentration was found to be within the normal therapeutic range. The dose was decreased and about five days later, her side effects disappeared. This case suggested that such side effects of digoxin as visual disturbance therefore need to be carfully monitored when ever digoxin and verapamil, which is one of the inhibitors of P-glycoprotein, are coadministered.

Theophylline infusion modulates prostaglandin and leukotriene production in man

Although theophylline has been used in the treatment of asthma for decades, it is not a first line choice any more. It is a well-known bronchodilator, but was recently discovered also to be an anti-inflammatory, immunomodulatory and bronchoprotective agent. Therefore we wanted to establish the role of theophylline on prostaglandin and leukotriene production, which plays a part in the pathogenesis of asthma. Theophylline was infused (bolus 5 mg/kg in 15 min and infusion 0.4 mg/kg/h for 1 h 45 min) into healthy volunteers. Thromboxane B 2, prostaglandin E 2 and leukotriene E 4 were measured from the A23187-stimulated whole blood samples and stable metabolites of thromboxane A 2; prostacyclin and leukotriene E 4 were measured from urine. Theophylline increased prostaglandin E 2 production and decreased leukotriene E 4 production ex vivo in whole blood, thus increasing the prostanoid/leukotriene ratio. It did not change thromboxane B 2 production stimulated by either spontaneous clotting or A23187 in the whole blood. Theophylline had hardly any effect on in vivo thromboxane, prostacyclin and leukotriene E 4 production measured as urinary metabolites, 11-dehydro-thromboxane B 2, 2,3-dinor-6-keto-prostaglandin F 1α and leukotriene E 4, respectively. Serum theophylline concentrations were at the lower level of normal therapeutic range during the infusion. The increase in PGE 2 and the decrease in LTE 4 synthesis ex vivo may offer a new explanation for the mode of antiasthmatic action of theophylline. It is notable that this phenomenon occurs at low serum theophylline concentrations. These results confirm the idea that theophylline has an anti-inflammatory and bronchoprotective action and support the use of theophylline as a therapeutic agent in asthmatic patients.

Drug administration in chronic liver disease.

Cirrhosis encompasses a range of pathophysiological changes that may alter drug disposition. Drugs that are dependent primarily on the liver for their systemic clearance are more likely to be subject to reduced elimination and subsequent accumulation. Drug accumulation may lead to excessive plasma drug concentrations and adverse effects, if the adverse effects of the drug are concentration-dependent. The effects of hepatic insufficiency on the pharmacokinetics of drugs are not consistent or predictable. Furthermore, the influence of hepatic disease on the disposition of various drugs can vary, even though those drugs may share the same apparent metabolic pathway. Problems in forecasting drug kinetic behaviour are further enhanced by the additional impairment of kidney function (frequently encountered in patients with advanced liver disease) and by the unpredictability of the glomerular filtration rate using customary methods in patients with cirrhosis. Accordingly, dosages are generally adapted empirically, with the help of serum drug concentrations, when applicable. However, drugs eliminated predominantly by hepatic metabolism are not among those most commonly inducing adverse drug reactions or causing severe complications in patients with cirrhosis. Electrolyte disturbances and the hepatorenal syndrome produced by furosemide (frusemide)-the disposition of which is not substantially modified in liver disease-appear to be the most frequent adverse drug reactions in patients with liver disease. Furthermore, clinically significant alterations in the action of medications at concentrations generally considered to be in the normal therapeutic range are not uncommon. Tissue responsiveness to the pharmacological action of some drugs may be modified, as evidenced by the increased susceptibility of the brain in patients with cirrhosis to the action of many psychoactive agents. Another example is the greater susceptibility of such patients to the nephrotoxic potential of aminogly-cosides, which should not be used in this patient group. Drugs may also interfere with adaptive physiological processes induced by liver disease. ACE inhibitors and nonsteroidal anti-inflammatory drugs counteract the enhanced activity of the renin-angiotensin system in advanced liver disease, thereby generating a high risk of excessive hypotension or acute renal failure, respectively. These drugs are best avoided in patients with cirrhosis. Finally, there may be pharmacological effects that overlap with some pathophysiological modifications related to the process of liver disease, such as increased portal pressure produced by some calcium antagonists, or hypoprothrombinaemia related to the inhibition of synthesis of vitamin K-dependent clotting factors by some beta-lactam antibacterials (especially moxalactam and cefamandole). Accordingly, the use of these drugs should be avoided in advanced liver disease. It is noteworthy that reduced drug metabolism in patients with liver disease does not seem to have a significant impact on the frequency of hepatotoxicity. Special caution should be exercised, however, in patients with alcoholic liver disease because excessive alcohol intake may potentiate the hepatotoxic effect of paracetamol (acetaminophen).

Effect of Acetylsalicylic Acid on Pulmonary Gas Exchange in Patients With Severe Pneumonia: A Pilot Study

It has been hypothesized that local release of prostacyclin in acute pneumonia may ablate hypoxic pulmonary vasoconstriction, thus contributing to the impairment of pulmonary gas exchange in these patients. Inhibition of cyclooxygenase pathway could prevent this phenomenon by reducing the release of these metabolites. A study was designed to assess the effect of I.V. acetylsalicylic acid (ASA) (2 g) on pulmonary gas exchange in seven patients (age, 64+/-11 [mean+/-SD] years) with unilateral severe pneumonia (PaO2/fraction of inspired oxygen, 168+/-67) needing mechanical ventilation. Respiratory gases, pulmonary and systemic hemodynamics, and ventilation-perfusion (VA/Q) distributions were studied before and 15 and 60 min after the infusion of ASA. At baseline, the amount of shunt (VA/Q ratios <0.005) was 28+/-17% of cardiac output, blood flow to areas with low VA/Q ratios (<0.1, excluding shunt) was 8+/-7%, and the dispersion of pulmonary blood flow distribution (second moment, log SD Q) was 1.45+/-0.49 (normal range, 0.3 to 0.6). Sixty minutes after the infusion of ASA, we observed a mild reduction of the amount of shunt, from 28+/-17% to 23.5+/-13% (p<0.05) without changes in arterial oxygenation. This was associated with a significant increase in mean pulmonary artery pressure (from 21.9+/-3.6 to 24.4+/-5.1 and 23.9+/-5.3 mm Hg, p<0.025 and p=0.1) and pulmonary vascular resistance (from 1.4+/-0.9 to 1.8+/-0.8 and 1.8+/-1.3 mm Hg x min x L(-1) , p<0.002 and p=0.11) 15 and 60 min after ASA, respectively. The ASA plasma levels were within the normal therapeutic range (120+/-7 microg/mL, 15 min, and 113+/-11 microg/mL, 60 min after ASA infusion). Although there was a modest improvement in intrapulmonary shunt, our results suggest that perfusion of ASA in this small sample of patients with severe pneumonia appears to be of little benefit as complementary treatment for severe hypoxemia.

Patterns of Craving and Pharmacokinetics in Long-Term Opiate Addicts in Methadone Maintenance Therapy

Results are presented and discussed of a clinical study of 20 long-term opiate addicts currently enrolled in Methadone Maintenance Therapy (MMT) and hospitalized in the Jellinek Clinic in Amsterdam. The purpose of this study was to better understand the craving and pharmacokinetics associated with difficulties in determining an optimal dosage schedule in MMT. Plasma methadone concentrations were determined using a newly developed HPLC (high pressure liquid chromatography) procedure. Drug craving was measured using the Experience Sampling Method (ESM), a new technique for measuring craving in MMT which provides data not previously reported in methadone literature. Within the subjects medicated in the normal therapeutic dosage range of MMT (10-90 mg/day), a significant relationship was found between higher methadone dose and higher craving levels, a finding that counters much of the accepted knowledge on MMT. Supporting this finding, some subjects showed high plasma methadone trough levels and high craving ...

A clinical assessment of the potential for pharmacological interaction between nimesulide and digoxin in patients with heart failure.

The potential interaction between nimesulide, a nonsteroidal anti-inflammatory drug, and digoxin was studied in 9 patients [6 males, 3 females; mean age 67 (range 57 to 70) years] with mild heart failure. All patients were receiving maintenance therapy with digoxin (0.25 mg/day, orally) and were treated with oral nimesulide 100mg twice daily for 7 days. Blood samples were collected at 8am and 6pm for 4 days before and throughout the nimesulide treatment period for determination of serum digoxin concentrations. Physical health, electrocardiographic recordings and blood and urine samples were also monitored. Mean serum digoxin concentrations remained within the normal therapeutic range throughout the study despite large interindividual variation. Furthermore, there were no significant differences between the morning and afternoon serum digoxin concentrations and there was no major change in the clinical condition of any patient. These results indicate that short term administration (7 days) of conventional therapeutic doses of nimesulide (100mg twice daily) does not modify the serum digoxin profile in patients with low class heart failure treated with a maintenance dose (0.25 mg/day) of this cardiac glycoside.

A critical reevaluation of the "therapeutic range" of aminoglycosides.

Routine pharmacokinetic drug monitoring has become an inherent component of aminoglycoside therapy over the last 10-15 years. The intent of this monitoring is to improve the outcome of treatment and to decrease the incidence of toxicity through the attainment and maintenance of serum aminoglycoside concentrations within a normal therapeutic range. The primary objective of this review was to critically analyze the scientific support for the following premises: (1) there is a causal relation between peak serum aminoglycoside concentrations in serum and the outcome of treatment; (2) there is a causal relation between trough serum aminoglycoside concentrations in serum and the outcome of treatment; (3) outcome is improved by monitoring and maintenance of serum aminoglycoside concentrations in the normal therapeutic range; (4) there is a causal relation between serum aminoglycoside concentrations and toxicity; and (5) monitoring and maintenance of serum aminoglycoside concentrations within a normal therapeutic range decrease the risk of toxicity. After a critical review of the literature, it was concluded that the evidence was insufficient to support the presently accepted normal therapeutic range. Recommendations for the monitoring of aminoglycoside therapy were drawn up.

Fatal Overdose of Amphotericin B in Two Preterm Infants

Two premature infants exhibited an acute cardiorespiratory collapse shortly after receiving a 50-fold overdose of amphotericin B (AmB). Despite aggressive intervention both infants died. Serum AmB levels were double the normal therapeutic range. We speculate that two less well-recognized toxic complications of AmB involving the cardiac and respiratory systems may have contributed to the fatal outcome.

Rapid high-performance liquid chromatographic methods for the determination of overdose concentrations of some non-steroidal anti-inflammatory drugs in plasma or serum

Separate methods are described for the determination of the non-steroidal anti-inflammatory drugs diflunisal, indomethacin, fenoprofen, ibuprofen, ketoprofen, naproxen, mefenamic acid and piroxicam at overdose concentrations in human plasma or serum, using high-performance liquid chromatography and ultraviolet detection. A common extraction, involving protein precipitation with acetonitrile, is employed for all methods. A 25 cm Hypersil ODS (5 μ particle size) analytical column is used for all chromatographic separations, with a mobile phase of acetonitrile-acetate buffer (pH 4.2 or 4.8). The methods are all reproducible and can also determine concentrations that fall within the normal therapeutic range for each drug.

Evaluation of Therapeutic Drug Monitoring in a Long-Term Care Facility: A Pilot Project

The need for a therapeutic drug monitoring service was evaluated in a 150-bed long-term care facility. Thirty blood samples from 28 residents (mean age 87.9 years) were assayed to determine trough drug concentrations. All subjects were examined to determine pharmacodynamic effect. Pharmacokinetic consultations were written for serum drug concentrations outside accepted ranges. Fifty percent (15 of 30) of serum drug levels measured were subtherapeutic; the remaining levels were in the normal therapeutic range. Based on this sample data, it could be concluded that a minimum of 32 percent and as many as 68 percent of serum drug levels would be subtherapeutic following drug analysis in similar nursing home populations. Of 12 consultations, recommendations for seven (58 percent) were accepted by the subject's primary care physicians. Four of the consultations (33 percent) resulted in dosage modifications. These results support the need for further study.

Effect of data collection method on results of serum digoxin concentration audit

The appropriateness of serum digoxin concentration (SDC) orders was evaluated with respect to indication for use, sampling time, and action taken by physicians when the reported SDC was out of the normal therapeutic range; the effect of the two data-collection methods used (retrospective and concurrent audits) on the results was studied. Criteria for the appropriate use of SDCs were approved by the medical staff through the pharmacy and therapeutics committee. Patients on adult medicine services were entered into the study as daily SDC determinations were reported by the clinical laboratory. Most of the SDCs were evaluated using approved criteria by primary pharmacist clinicians who were concurrently monitoring drug therapy and participating with the treatment team. A retrospective audit of the same patients was conducted, using only chart review. A total of 134 SDCs involving 78 patients were evaluated. Concurrent-audit results indicated that 18.7% of the SDCs were ordered without an appropriate indication, 16.4% were sampled incorrectly with respect to proper timing, and 8.2% did not result in dosage adjustments when indicated. With respect to appropriate sampling time and overall use of SDCs, significantly more SDCs met the standards under concurrent audit than under retrospective audit. The retrospective chart review method of auditing may not detect as much pertinent information as is desirable.

Effects of antidepressants on entrainment of circadian rhythms

Brown Gregory H. and Jo Seggie: Effects of antidepressants on entrainment of circadian rhythms. Prog. Neuro-Psychopharmacol. & Biol. Psychiat. 1988, 12 :299–306 1. 1. Anti-depressant drugs alter the period of circadian rhythms under free running conditions. We have examined whether treatment with anti-depressants alters such rhythms under conditions of entrainment to a lighting cycle or following a phase advance of the lighting. 2. 2. In the first experiment, following entrainment to a 10L:14D lighting schedule, male Long Evans rats were injected daily for three weeks, with either saline, desipramine or clomipramine at a dose of 10 mg/kg. Under these conditions, the resting circadian patterns of serum corticosterone and melatonin were not affected. 3. 3. In a second experiment, male Wistar rats, adapted to a 12L:12D lighting cycle, were implanted with transmitters, permitting continuous recording of temperature. Animals were given chow supplemented with clomipramine or desipramine so that plasma clomipramine levels were within the normal therapeutic range for humans and desipramine levels were at 4–6 times the therapeutic range. After three weeks of drug treatment, the mean 24 h pattern of temperature was decreased and phase advanced by 72 min in the desipramine group compared to clomipramine or controls. Following a 6 h phase advance in the lighting cycle, the three groups responded differently. The clomipramine group evidenced the most rapid advance in the acrophase, equivalent to 6.9 h, while the saline and desipramine treated groups re-entrained significantly more slowly. 4. 4. Clomipramine, given to achieve clinically meaningful blood levels, does not alter the circadian temperature rhythm under constant conditions of alternating light and dark, but significantly accelerates the rate of re-entrainment to a 6 h phase advance of the lighting schedule.

Approche physiopathologique des hyperlactatémies pathologiques chez le diabétique. Intérêt de la metforminémie

Type B lactic acidosis, or pathological hyperlactataemia (PHL), is defined by an arterial lactate level greater than 5 mmol · 1 −1. It is a known and severe complication of diabetes mellitus treated with biguanide hypoglycaemic agents, particularly phenformin which was taken off the French pharmaceutical market in 1977. Metformin, which remains the only biguanide hypoglycaemic agent currently prescribed in France, may also lead to this complication. However it does so less frequently and mostly in the diabetic presenting with renal failure. A few well studied cases showed that PHL could be correlated with excessive metformin blood levels, i.e. a toxic mechanism. In order to find out whether this toxic mechanism was the real cause of PHL in diabetics treated with metformin, a systematic study of metformin blood levels was carried out in 20 such patients. They had all been admitted to a critical care unit presenting with PHL. The results of this study led us to distinguish between two groups of patients. The seven patients of the first group had high metformin blood levels (4.3 to 65.8 g · 1 −1). In these, renal excretion or extrarenal dialysis lowered or normalized their hyperlactatemia, and six of the seven recovered from PHL. In the second group, with thirteen patients, metformin blood levels were within the normal therapeutic range (0.225 to 3 g · 1 −1) for seven patients and close to zero for the other six. This second group received the same treatment as the first one. Only three patients recovered, the others all died. This study as well as that of 16 cases from the literature appeared to show that PHL in diabetic patients treated with metformin was the result of two different pathological mechanisms. On one hand, there seemed to be toxic accumulation of this biguanide, favoured by renal failure. On the other hand, severe anoxaemic aggressions seemed to trigger PHL in some susceptible patients treated with metformin (group II) and in others not receiving biguanides (group III, 8 cases of diabetic patients not treated with metformin). This second mechanism could trigger PHL in all diabetics, probably independently of biguanide intake. The essential difference between these two types of PHL was prognosis : that of PHL due to metformin accumulation appeared to be correlated with its excretion, and was quite good; however, that of PHL due to an anoxaemic mechanism was rather poor.