Serum Diazepam Concentrations in Overdose Their Significance PETER JATLOW, M. D., KENNETH DOBULAR, M.D. AND DAVID BAILEY, M.D.
DIAZEPAM IS REPORTED to be the most widely prescribed drug in the United States. Statistically, therefore, one would expect it to be commonly implicated in drug intoxications, and indeed, a high incidence of apparent overdose with this compound is seen in hospital emergency rooms. 9 1 7 1 8 The general impression that overdoses with diazepam or other benzodiazepines are unlikely, by themselves, to be life-threatening 31 ' has recently been supported by a large clinical series recently reported by Greenblatt and associates. 9 Considering the unreliability of clinical history in such situations, however, without drug levels it is difficult to be certain that a quantitatively significant acute ingestion has occurred. Received July 31, 1978; accepted for publication August 24, 1978. Supported in part by USPHS Grant NIDA R01 DA 01294. Address reprint requests to Dr. Jatlow: Department of Laboratory Medicine, Yale University School of Medicine, 333 Cedar Street, New Haven, Connecticut 06510. 0002-9173/79/1000/0571 $00.85 © Ai 571
Department of Laboratory Medicine, Yale University School of Medicine, New Haven, Connecticut
Furthermore, since drug levels are frequently requested and are becoming increasingly available, it is important to know what, if any, significance should be attached to drug concentrations in the emergency room management of the patient with possible diazepam overdose. In addition, accurate diagnosis of overdose is important for the correct long-term follow-up care of the patient, regardless of whether the particular acute episode is life-threatening. To date no large series of cases of diazepam overdose, with blood levels, is available. We have correlated total serum benzodiazepine concentrations in 93 cases of diazepam overdose and in five cases of chlorazepate overdose with clinical manifestations. In addition, separate diazepam and nordiazepam concentrations have been determined by gas chromatography with nitrogen detector in specimens from 101 cases of diazepam overdose. Materials and Methods A total of 184 cases of drug overdose involving diazepam ingestion were seen in the Yale-New Haven Hospital Emergency Room* and documented by chemical analysis between October 1968 and July 1975. A sufficiently detailed emergency room record, and hospital chart when the patient was admitted, were available for review in 93 cases, as well as in five cases of clorazepate overdose. All clinical data were obtained by retrospective review of these records. Coma grading essentially followed the classification of Reed, lli and was as follows: A: Awake Grade 0: Asleep, but can be aroused Grade I: Comatose, but responsive to painful stimuli; no respiratory or circulatory depression; reflexes intact * A small percentage of these (5%) involved specimens sent from the West Haven VA Hospital and other Connecticut hospitals. ican Society of Clinical Pathologists
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 5, 2016
Jatlow, Peter, Dobular, Kenneth, and Bailey, David. Serum diazepam concentrations in overdose. Their significance. Am J Clin Pathol 72: 571-577, 1979. Total serum benzodiazepine concentrations were correlated with clinical manifestations in 93 cases of diazepam overdose. Diazepam and nordiazepam were also each separately determined in 101 serum specimens from cases of diazepam overdose, including 27 cases from the aforementioned clinical correlation study. In addition, serum nordiazepam concentrations were measured in five cases of clorazepate overdose. Concentrations of total benzodiazepine ranged from 1 to 22 fig/ml. All patients survived with supportive therapy only. Each of the 25 patients who had ingested only diazepam was awake or in grade 0 coma, even when drug concentrations were ten-fold greater than the accepted upper limit of the therapeutic range. None of the patients who had ingested only diazepam needed hospitalization; all were discharged from acute medical care after a period of emergency room observation. The ratios of parent drug to N-desmethyl metabolite (nordiazepam) in those overdose specimens analyzed by gas chromatography averaged 3:1. This high ratio may be useful in differentiating acute overdose from high concentrations resulting from chronic therapy. Although determination of diazepam concentrations aid in establishing that an overdose has occurred, when more than grade I or II coma is present, other drugs or an alternative explanation should be sought, regardless of the drug concentration. (Key words: Diazepam; Nordiazepam; Overdose.)
572
A.J.C.P. • October 1979
JATLOW, DOBULAR, AND BAILEY
Grade II: Does not withdraw from painful stimuli; no respiratory or circulatory depression; most reflexes intact Grade III: Most or all reflexes absent; no respiratory or circulatory depression Grade IV: Most or all reflexes absent; significant respiratory or circulatory depression, or both
To assess the specificity of the spectrophotometric assay, and to determine selectively the concentrations of parent drug and metabolite, some specimens were also analyzed by gas chromatography using a nitrogen detector. During the latter part of the study, when
Chromatography was performed on a Perkin-Elmer (Norwalk, Conn.) gas chromatograph (Model 3920) equipped with a nitrogen detector. Gas chromatographic conditions were: column, 3% OV-17 on Gas Chrom Q: (Applied Science), oven 255 C, isothermal; injector 270 C; detector, 300 C, helium, 20 ml/min. The bead temperature was arbitrarily adjusted each day to provide adequate and comparable sensitivity. Under these conditions the retention times of diazepam, nordiazepam and madazepam were 8.3, 11.8, and 4.2 min, respectively. The within-run and between-run coefficients of variation were 4.6% and 4.8%, respectively, for diazepam, and 5.4% and 6.8% for nordiazepam. Although this procedure may be sufficiently sensitive for therapeutic monitoring, it was evaluated and optimized for use in the overdose range (>1 /ug/ml) only. Various common basic drugs that might have been coextracted did not interfere, including methaqualone, imipramine, despramine, amitriptyline, nortriptyline, methapyriline, caffeine, and nicotine. Results Single-drug (Diazepam only)
Ingestion
On the basis of histories or drug analysis, or both diazepam was the only drug ingested in 38 (41%) of the
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 5, 2016
At the time of admission to the emergency room, every patient received either an "emergency toxicology screen" or only a serum diazepam analysis. The routinely used toxicology screen, which has been published, 14 is offered 24 hours a day, and includes the following: an ultraviolet spectrophotometric analysis of 5,5' disubstituted barbiturates, chlordiazepoxide, methaqualone, diazepam, and glutethimide; qualitative urine spot tests for imipramine and phenothiazines; a blood alcohol determination. In most instances, thin-layer chromatography of urine was subsequently performed. Specific identifications of barbiturates and alcohols were performed as indicated. The routine spectrophotometric diazepam analysis performed at the time the patient was in the emergency room was as follows: a 3-ml volume of serum was buffered topH 7.4 with 0.5 ml 0.5 M phosphate buffer, and extracted with 30 ml chloroform. A 25-ml aliquot of the chloroform was washed successively with 0.45 M NaOH (discarded or used for barbiturate analysis) and 0.5 M HCl (discarded or saved for chlordiazepoxide assay). The chloroform was evaporated, the residue was dissolved in 5 ml of hexane, and the diazepam (and nordiazepam) was extracted into 4 ml 2 M HCl. The aqueous acid phase was scanned in an ultraviolet recording spectrophotometer, "diazepam" identified by its spectrum and the concentration determined by comparing the absorbance difference between 280 and 300 nm with that of similarly processed serum standards. This procedure has a coefficient of variation of 4% and is linear over a range of 1 to 20 ^ig/ml. It does not differentiate among diazepam, nordiazepam, and oxazepam, nor is it intended to be sufficiently sensitive for therapeutic monitoring. However, chlordiazepoxide, and other bases with higher/?K a s that are generally present at much lower concentrations, are removed from the chloroform with the earlier HCl wash and do not interfere. Methaqualone and its metabolites will interfere but can be recognized by their spectra. The concentrations obtained by this method, in routine use at the time of admission, were used for clinical correlation.
sufficient serum remained after spectrophotometric analysis, it was frozen for future gas chromatographic assay. In order to obtain a larger sampling for the analytic comparison, 74 additional specimens from diazepam overdose cases, from patients seen after July 1975, were also reassayed by gas chromatography but not used in the clinical study. Nordiazepam concentrations following clorazepate overdose were all determined by gas chromatography. Gas chromatography was performed as follows: 0.5 ml of plasma was alkalinized with 1 ml saturated Na 3 P0 4 8 and extracted with 8 ml hexane containing 5% acetone (V/V) and 2.5 /xg medazepam as an internal standard. After centrifugation, the solvent was transferred, and back-extracted with 1 ml 0.1 M H 2 S0 4 . After centrifugation the aqueous phase was separated, alkalinized with saturated Na 2 C0 3 , and re-extracted with 5 ml hexane:acetone (without internal standard). The solvent was separated, evaporated under a stream of nitrogen, the residue dissolved in 20 fi\ methanol, and 2 - 4 fi\ submitted to chromatography. Concentrations were determined by comparison of relative peak areas with those of extracted plasma standards. Plasma standards supplemented with exact concentrations of diazepam and nordiazepam were carried through the entire procedure.
DIAZEPAM CONCENTRATIONS IN OVERDOSE
Vol. 72 • No. 4
573
Table I. Diazepam and Ethanol Ingestion
Case
300 100-150 60 40-200
9 300 500 30 75 30 300 35 200 •) 50 200 100 7 150 90 60 1,000 7 65 50-75 7 150 80
Other Drugs
Coma Grade
Ethanol Concentration (mg/dl)
Total Benzodiazepine Concentration (/xg/ml)
Caffeine; ergotamine — — — Chlordiazepoxide (8 /ig/ml) — — — 1,000 mg amitriptyline — — Amitriptyline: perphenazine — — — — 50 mg doxepin Propoxyphene — — — — — Diphenylhydantoin (25 /xg/ml) Flurazepam — — Secobarbital (0.4 mg/dl) — —
0 0 0 0 A 0 A A A 0 0 0 0 0 0 A 0 A 0 III A 0 0 0 0 A 0 A IV 0 0
175 325 150 150 181 225 190 180 78 148 143 273 80 110 198 205 159 160 109 130 216 53 170 160 204 192 220 215 70 128 104
2 22 4 3 2.5 2.5 4 3 11 3 5 2 2 8 2 7 4 2 5 4 6 3 2 4 6 2 2 2 3 3 3
93 cases. The total benzodiazepine concentrations in this group ranged from 1 to 9 /u,g/ml. Estimated doses (by history) ranged as high as 600 mg. Sixteen patients were awake, and 22 were in grade 0 coma. The mean drug concentration of those who were awake (3.7 /Ag/ml) did not differ from that of those classified as in grade 0 coma (3.7 /u.g/ml). None of the patients who had only ingested diazepam needed medical hospitalization, regardless of drug level or doses; all were either released or transferred to psychiatric facilities after a period of observation in the emergency room.
grade 0 (asleep but arousable) and A (awake) probably reflected tolerance rather than toxicity.
Diazepam with Ethanol
Grade IV Coma
Concurrent ethanol ingestion was established by chemical analysis in 31 cases, and was the only other drug ingested in 21 of them. Table 1 summarizes the data from the 30 cases in which ethanol concentrations were greater than 50 mg/dl. Twenty-nine of the 31 patients in this group were either awake (9) or in grade 0 coma. These included four patients with diazepam levels of 6 /xg/ml or more in association with ethanol concentrations greater than 200 mg/dl. Since many of these patients were chronic drinkers, and were receiving diazepam chronically, the difference between
Seven patients were classified as in grade IV coma. All needed assisted ventilation. Total serum benzodiazepine concentrations in this group ranged from 3 to 7 £tg/ml. All of the patients in grade IV coma had ingested other drugs in addition to diazepam. In fact, as shown in Table 2, six of the seven had ingested two or more additional drugs.
Multi-drug
Ingestion
Fifty-six (61%) of the patients had ingested other drugs with diazepam. In 35 instances, additional drugs other than ethanol were also taken. Most common were short acting-barbiturates (nine cases), tricyclic antidepressants (seven cases) and phenothiazines (six cases). A drug classifiable in one of these three groups was implicated in every case with grade IV coma.
Analytic
Considerations
Comparison of the benzodiazepine concentrations determined by scanning ultraviolet spectrophotometry
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 5, 2016
5 6 10 14 15 18 19 20 21 23 26 30 31 32 36 47 49 52 53 63 67 71 76 78 79 83 87 91 92 93 95
Purported Diazepam Dose (mg)
A.J.C.P. • October 1979
JATLOW, DOBULAR, AND BAILEY
574
B.
LJ O
CD CC
° 0.8 h < 1.0 1.2 .4 200
J_ _1_ 250 300 WAVELENGTH (nm)
350
4 8 TIME (min)
12
Fio. \. A, ultraviolet absorption spectrum from serum of a patient following diazepam overdose. B, gas chromatogram from serum of the same patient, a = medazepam (internal standard); b = diazepam; c = nordiazepam. with the sum of diazepam and nordiazepam measured by gas chromatography yielded a correlation coefficient of 0.87. The mean total benzodiazepine concentrations of the 101 samples determined by ultraviolet spectrophotometry and gas chromatography were 4.97 jttg/ml and 4.70 ^g/ml, respectively. The means did not differ significantly (t test for paired observations). The ultraviolet spectrum and gas chromatographic pattern from a patient's serum are shown in Figure 1.
Parent Drug.Metabolite
Ratios
These were determined for 101 emergency-room specimens by gas chromatography, as previously described. In almost all instances the concentration of parent drug was considerably higher than that of the N-desmethyl metabolite. The ratios of parent drug to metabolite averaged 3.0. The data for those 27 patients for whom specific parent drug and metabolite concentrations were determined and clinical histories reviewed are summarized in Table 3.
Table 2. Diazepam Overdose with Grade IV Coma
Case
Diazepam Dose (mg)
16 17 25 46 30 55 92
400 ? 500 450 9 75 9
Other Drugs
Coma Grade
Total Benzodiazepine Concentration (/i.g/ml)
Imipramine, prochlorperazine propoxyphene, theophylline, ephedrine Codeine, ethchlorvynol, lithium (9.5 mg/1) 1,200 mg amitriptyline, 250 mg chlorpromazine Diphenylhydantoin, amitriptyline, propoxyphene Secobarbital (1.3 mg/dl) EthanOl (200 mg/dl), 120 mg perphenazine Secobarbital (0.4 mg/dl), ethanol (70 mg/dl)
IV IV IV IV IV IV IV
7 4 5 6 5 5 3
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 5, 2016
z 0.6
Ethanol, 180 mg/dl Barbiturate, 0.5 mg/dl; imipramine, prochlorperazine; propoxyphene Codeine, ethchlorvynol Ethanol. 180 mg/dl Ethanol, 78 mg/dl; umitriptyline Ethanol, 148 mg/dl; amitriptyline, codeine
400
150-300 500 7 500 300
1,200 mg amitriptyline; 250 mg chlorpromazine Ethanol, 110 mg/dl
60-600 30 60
2-3 g glutethimide
250 50-120
Ethanol, 205 mg/dl Secobarbital, 1.3 mg/dl Glutethimide, 0.4 mg/dl 120 mg tranylcypromine
200-250 7 7 250-500 600 7 7
Pseudoephedrine, prochlorperazine 250-500 mg chlordiazepoxide
500
Data from five cases of clorazepate overdose are shown in Table 4. These patients, none of whom required hospitalization, has serum nordiazepam concentrations ranging from 0.7 to 5.1 /xg/ml. Discussion The immediate management of possible drug overdose is, of course, supportive, with a few exceptions not specific, and relatively independent of drug identification or quantitation. An accurate drug analysis is valuable for confirmation, decisions relating to disposition, further diagnostic work-up of coma, establishing prognosis, and long-term follow-up. When confronted with a probable drug overdose, the emergency room physician will nonetheless often request a stat drug analysis, regardless of the availa-
Coma Grade
Diazepam (HSlml)
Nordiazepam (/ig/ml)
I'otal Drug (jig/ml)
0 0 0 0 0 A
3.1 1.1 0.9 1.6 8.4 0.3
0.1 0.0 0.4 0.7 0.3 1.7
3.2 1.1 1.3 2.3 8.7 2.0
IV IV A A 0 0 IV 0 0 0 A 0 0 A IV A A A A 0 A
6.2 1.0 2.8 8.5 7.8 3.9 2.7 7.5 6.6 2.6 2.6 4.4 3.1 4.2 5.0 1.8 6.0 4.2 5.4 2.8 5.0
0.6 1.7 1.1 1.2 0.0 0.2 0.2 2.4 0.8 1.7 0.3 0.4 0.1 0.2 0.9 0.3 0.6 0.4 0.6 0.5 0.6
6.8 2.7 3.9 9.7 7.8 4.1 2.9 9.9 7.4 4.3 2.9 4.8 3.2 4.6 5.9 2.1 6.6 4.6 6.0 3.3 5.6
bility of information necessary for meaninful interpretation of these data. Since diazepam reaches levels greater than 1 - 2 ptg/ml following overdose, it can under such conditions be rapidly measured by such relatively uncomplex technics as ultraviolet spectrophotometry, fluorimetry, or even gas chromatography with conventional (flame ionization) detectors. As a consequence, drug concentrations are more commonly available following diazepam overdose than, for example, in the more serious instance of tricyclic antidepressant intoxication. Since the described spectrophotometric analysis of diazepam is available at all times (24 hours/day) for cases of suspected overdose, these data were available and used for clinical correlation studies on the 93 cases. Because we wished to validate the data for use in
Table 4. Nordiazepam Concentrations Following Clorazepate Overdose Case la 2a 3 4
Purported Dose (mg) 23 tablets (? mg) 22.5 4-6 tablets (? mg) 50-300
Other Drugs Ethanol, 125 mg/dl Barbiturate (amobarbital and secobarbital), I mg/dl; ethanol, 43 mg/dl; codeine, amitriptyline Ethanol, 253 mg/dl
Coma Grade
Nordiazepam (/xg/ml)
Total Drug (/ug/ml)
A A A
5.1 0.7 2.0
5.1 0.7 2.0
II A
1.0 2.0
1.0 2.0
Diazepam (eg/ml)
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 5, 2016
17 20 21 23 24 25 32 39 40 41 43 45 47 50 51 77 80 82 85 86
Purported Dose (mg)
576
JATLOW, DOBULAR, AND BAILEY
yses was to validate the spectrophotometric data used for clinical correlation, so as to establish its suitability for emergency toxicology. The good correlation between the spectrophotometric values and the sum of diazepam and nordiazepam as measured by gas chromatography indicates that the ultraviolet spectrophotometric procedure provides a sufficiently accurate measurement of total active compound for clinical evaluation of suspected overdose. While oxazepam, the major urinary metabolite of diazepam, is also pharmacologically active, it does not reach significant serum concentrations. Spectrophotometric procedures can thus be safely recommended as a valid means to confirm that a diazepam "overdose" (excess ingestion) has occurred. It is not nearly sensitive enough for use in the therapeutic range, nor does it differentiate between diazepam and its major active metabolites. The gas chromatographic analyses were also performed to establish the relative concentrations of diazepam and its active N-desmethyl metabolite following overdose. Following chronic therapy, concentrations of nordiazepam are usually comparable to or exceed that of the parent compound. In this overdose series, concentrations of parent drug were generally well in excess of the metabolite, with an average parent to metabolite ratio of 3:1. While not surprising, this information may be useful in determining whether a high drug concentration, and perhaps mild clinical toxicity, is a result of chronic high-dose therapy or an acute overdose. Following chlorazepate overdose, only its major active derivative nordiazepam was measured. Although the number of cases is small, the data suggest that the consequences of clorazepate overdose may be no more serious than those of diazepam ingestion. These data lead to the question whether drug measurements are of any value in managing possible diazepam overdose. Experienced clinical impressions 10 and recent clinical reviews of benzodiazepine overdose 9 1 8 (without drug levels) have strongly suggested that benzodiazepine overdose is relatively benign. Nonetheless, various tabulations 120 and available manufacturer's product descriptions have indicated that serious if not fatal coma is a consequence of diazepam intoxication. Our studies indicate that diazepam overdose, confirmed by high serum drug concentrations, is not likely to produce deep coma (greater than grade I) or to necessitate hospital admission, regardless of the dose or drug level. However, lesser manifestations of central nervous system depression, such as somnolence, which also must be evaluated by the emergency room physician, can be attributed to an elevated diazepam concentration. Furthermore, despite the lack of alarming symptoms, correct diagnosis
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 5, 2016
clinical correlation studies and also to assess the relative contributions of parent drug and metabolite, a separate gas chromatographic analysis was performed. Since only 27 specimens from the original clinical correlation study were available for reanalysis, specimens from an additional 74 cases were also analyzed by both ultraviolet spectrophotometry and gas chromatography with a nitrogen detector. A retrospective clinical review for classification of central nervous system depression was not performed in this latter group of 74 cases. The conclusions to be drawn from this study are almost self-explanatory. Large overdoses with diazepam and high blood drug concentrations are generally not life-threatening, nor are they likely to produce coma. None of the patients who had ingested only diazepam were in deeper than grade 0 coma, despite serum drug concentrations well in excess of those seen following therapy. Following chronic doses of 30 mg/day, total benzodiazepine concentrations slightly more than 1 fig/ ml have been reported, 7 although daily doses of 10 mg/ day produce concentrations of less than 1 /ug/ml.7,15 The therapeutic plasma concentrations requested for adequate antianxiety effect are also reported to be < 1 jig/ml. 613 Massive chronic daily doses of 150-200 mg in one subject were reported to yield total benzodiazepine concentrations of 5-6 fx.g/m\.7 Published tables of drug levels have listed 5-20 /ug/ml as toxic and more than 20 /ng/ml as lethal. 1,20 Many of the patients in this series were in the so-called toxic range, yet did not manifest sufficient central nervous system depression to need hospitalization. Ethanol, despite its known positive interaction with many central nervous system depressants, and its reported enhancement of diazepam absorption, 12 appeared to have little impact on the sfeVerity of diazepam overdose. Concentrations of ethanol of more than 200 mg/dl in association with diazepam concentrations well above the therapeutic range were not associated with toxicity. Many of these patients did have histories of alcohol abuse, and may have had increased tolerance for central nervous system depressants. Mixed ingestions (other than ethanol) were most commonly associated with barbiturates, pheriothiazines, and tricyclic antidepressants, all of which are recognized to cause serious intoxication. Indeed, all of those patients in grade IV coma had ingested one or more of these drugs. Only case 92 is somewhat difficult to reconcile with the ingestion history, and laboratory data. In the others the clinical manifestations (Table 2) could reasonably be attributed to the other additional drug(s), regardless of the diazepam level. One reason for the large-scale comparison between the spectrophotometric and gas chromatographic anal-
A.J.C.P. • October 1979
Vol. 72 • No. 4
DIAZEPAM CONCENTRATIONS IN OVERDOSE
2. 3. 4. 5. 6.
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 5, 2016
centrations of diazepam and its metabolite N desmethyldiazepam in relation to anxiolytic effect. Clin Pharmacol Ther 15:473-483, 1974 7. DeSilva JA, Koechlin BA, Bader B: Blood level distribution patterns of diazepam and its major metabolite in man. J Pharm Sci 55:692-702, 1977 8. DeSilva JAF, Bekersky I, Puglisi CV, et al: Determination of 1,4-benzodiazepines and diazepine-2-ones in blood by electron capture gas chromatography. Anal Chem 48:10-19. 1976 9. Greenblatt DJ, Allen MD. Noel BJ, et al: Acute overdose with benzodiazepine derivatives. Clin Pharmacol Ther 21:497-514. 1977 10. Greenblatt DJ, Shader RI: Psychotropic drug overdose. Manual of Psychiatric Therapeutics. Edited by RI Shader. Boston. Little. Brown, 1975, pp 237-268 11. Greenblatt DJ, Shader RI: Benzodiazepines in Clinical Practice. New York Raven Press, 1974 12. Hayes SL, Pablo G, Radomski T. et al: Ethanol and oral diazepam absorption. N Engl J Med 296:186-189, 1977 13. Hillestad L, Hansen T, Melsom H, et al: Diazepam metabolism in normal man. I. Serum concentrations and clinical effects after intravenous, intramuscular and oral administration. Clin Pharmacol Ther 16:479-487. 1974 14. Jatlow P: UV spectrophotometry for sedative drugs frequently involved in overdose emergencies. Methodology for Analytical Toxicology. Edited by I Sunshine. Cleveland. Ohio Chem Rubber Co., 1974. pp 414-420 15. Kaplan SA, Jack ML. Alexander K, et al: Pharmacokinetic profile of diazepam in man following single intravenous and oral and chronic oral administrations. J Pharm Sci 62:1789-1976. 1973 16. Reed CE, Driggs MF. Foote CC: Acute barbiturate intoxication: References A study of 300 cases based on a physiologic system of clasBaselt RC, Wright JA, Cravey RH: Therapeutic and toxic consification of the severity of the intoxication. Ann Intern Med centrations of more than 100 toxicologically significant drugs 37:290-303, 1952 in blood, plasma, or serum: A tabulation. Clin Chem 21:44-62, 17. Regent TA, Wahl KC: Diazepam abuse: Incidence, rapid screen1975 ing and confirming methods. Clin Chem 22:889-891. 1976 Bell EF: The use of naloxone in the treatment of diazepam 18. Roche Scientific Report: Incidence of fatalities associated with poisoning. J Pediatr 87:803-804, 1975 Valium (diazepam) alone or in combination with other subBenzodiazepine overdose. Med Lett Drugs Ther 18:60, 1976 stances. Hoffman-LaRoche Inc, Nutley, N. J.. 1976 Berger R, Green B, Mclnick A: Cardiac arrest caused by oral 19. Sparks H, Goldman AS: Diazepam intoxication in a child. Am J diazepam intoxication. Clin Pediatr 14:842-844, 1977 Dis Child 109:128-129. 1965 Cate JC, Jatlow PI: Chlordiazepoxide overdose. Interpretation 20. Winek CL: Tabulation of therapeutic, toxics and lethal concenof serum drug concentrations. Clin Toxicol 6:553-561, 1973 trations of drugs and chemicals in blood. Clin Chem 22:832— Dasberg HH, van der Kleign E, Guelen PJR, et al: Plasma con836. 1976
of diazepam overdose is, of course, important for longterm follow-up, including psychotherapy, particularly in prevention of a second attempt with, perhaps, a more potent compound. Following our earlier study of another benzodiazepine, chlordiazepoxide (Librium), we reached similar conclusions about overdose with that drug."' In summary, whether a diagnosis of diazepam overdose is based upon history or drug measurement, or both, when coma more severe than grade I is present, another drug or explanation should be sought. There will, in all probability, be exceptions to this generalization among susceptible individuals (e.g., patients with myxedema, those with chronic pulmonary disease, the very elderly and the very young). Serious diazepam overdose with coma has been reported to occur in small children. 2 4 1 9 Drug measurements are still useful in establishing that an acute ingestion has occurred, and to explain mild central nervous system depression. Despite the suggested utility of diazepam measurements, our data suggest that there is rarely a justifiable need for an ultrastat or desperate emergency serum diazepam analysis. 1.
577
DIAZEPAM IS REPORTED to be the most widely prescribed drug in the United States. Statistically, therefore, one would expect it to be commonly implicated in drug intoxications, and indeed, a high incidence of apparent overdose with this compound is seen in hospital emergency rooms. 9 1 7 1 8 The general impression that overdoses with diazepam or other benzodiazepines are unlikely, by themselves, to be life-threatening 31 ' has recently been supported by a large clinical series recently reported by Greenblatt and associates. 9 Considering the unreliability of clinical history in such situations, however, without drug levels it is difficult to be certain that a quantitatively significant acute ingestion has occurred. Received July 31, 1978; accepted for publication August 24, 1978. Supported in part by USPHS Grant NIDA R01 DA 01294. Address reprint requests to Dr. Jatlow: Department of Laboratory Medicine, Yale University School of Medicine, 333 Cedar Street, New Haven, Connecticut 06510. 0002-9173/79/1000/0571 $00.85 © Ai 571
Department of Laboratory Medicine, Yale University School of Medicine, New Haven, Connecticut
Furthermore, since drug levels are frequently requested and are becoming increasingly available, it is important to know what, if any, significance should be attached to drug concentrations in the emergency room management of the patient with possible diazepam overdose. In addition, accurate diagnosis of overdose is important for the correct long-term follow-up care of the patient, regardless of whether the particular acute episode is life-threatening. To date no large series of cases of diazepam overdose, with blood levels, is available. We have correlated total serum benzodiazepine concentrations in 93 cases of diazepam overdose and in five cases of chlorazepate overdose with clinical manifestations. In addition, separate diazepam and nordiazepam concentrations have been determined by gas chromatography with nitrogen detector in specimens from 101 cases of diazepam overdose. Materials and Methods A total of 184 cases of drug overdose involving diazepam ingestion were seen in the Yale-New Haven Hospital Emergency Room* and documented by chemical analysis between October 1968 and July 1975. A sufficiently detailed emergency room record, and hospital chart when the patient was admitted, were available for review in 93 cases, as well as in five cases of clorazepate overdose. All clinical data were obtained by retrospective review of these records. Coma grading essentially followed the classification of Reed, lli and was as follows: A: Awake Grade 0: Asleep, but can be aroused Grade I: Comatose, but responsive to painful stimuli; no respiratory or circulatory depression; reflexes intact * A small percentage of these (5%) involved specimens sent from the West Haven VA Hospital and other Connecticut hospitals. ican Society of Clinical Pathologists
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 5, 2016
Jatlow, Peter, Dobular, Kenneth, and Bailey, David. Serum diazepam concentrations in overdose. Their significance. Am J Clin Pathol 72: 571-577, 1979. Total serum benzodiazepine concentrations were correlated with clinical manifestations in 93 cases of diazepam overdose. Diazepam and nordiazepam were also each separately determined in 101 serum specimens from cases of diazepam overdose, including 27 cases from the aforementioned clinical correlation study. In addition, serum nordiazepam concentrations were measured in five cases of clorazepate overdose. Concentrations of total benzodiazepine ranged from 1 to 22 fig/ml. All patients survived with supportive therapy only. Each of the 25 patients who had ingested only diazepam was awake or in grade 0 coma, even when drug concentrations were ten-fold greater than the accepted upper limit of the therapeutic range. None of the patients who had ingested only diazepam needed hospitalization; all were discharged from acute medical care after a period of emergency room observation. The ratios of parent drug to N-desmethyl metabolite (nordiazepam) in those overdose specimens analyzed by gas chromatography averaged 3:1. This high ratio may be useful in differentiating acute overdose from high concentrations resulting from chronic therapy. Although determination of diazepam concentrations aid in establishing that an overdose has occurred, when more than grade I or II coma is present, other drugs or an alternative explanation should be sought, regardless of the drug concentration. (Key words: Diazepam; Nordiazepam; Overdose.)
572
A.J.C.P. • October 1979
JATLOW, DOBULAR, AND BAILEY
Grade II: Does not withdraw from painful stimuli; no respiratory or circulatory depression; most reflexes intact Grade III: Most or all reflexes absent; no respiratory or circulatory depression Grade IV: Most or all reflexes absent; significant respiratory or circulatory depression, or both
To assess the specificity of the spectrophotometric assay, and to determine selectively the concentrations of parent drug and metabolite, some specimens were also analyzed by gas chromatography using a nitrogen detector. During the latter part of the study, when
Chromatography was performed on a Perkin-Elmer (Norwalk, Conn.) gas chromatograph (Model 3920) equipped with a nitrogen detector. Gas chromatographic conditions were: column, 3% OV-17 on Gas Chrom Q: (Applied Science), oven 255 C, isothermal; injector 270 C; detector, 300 C, helium, 20 ml/min. The bead temperature was arbitrarily adjusted each day to provide adequate and comparable sensitivity. Under these conditions the retention times of diazepam, nordiazepam and madazepam were 8.3, 11.8, and 4.2 min, respectively. The within-run and between-run coefficients of variation were 4.6% and 4.8%, respectively, for diazepam, and 5.4% and 6.8% for nordiazepam. Although this procedure may be sufficiently sensitive for therapeutic monitoring, it was evaluated and optimized for use in the overdose range (>1 /ug/ml) only. Various common basic drugs that might have been coextracted did not interfere, including methaqualone, imipramine, despramine, amitriptyline, nortriptyline, methapyriline, caffeine, and nicotine. Results Single-drug (Diazepam only)
Ingestion
On the basis of histories or drug analysis, or both diazepam was the only drug ingested in 38 (41%) of the
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 5, 2016
At the time of admission to the emergency room, every patient received either an "emergency toxicology screen" or only a serum diazepam analysis. The routinely used toxicology screen, which has been published, 14 is offered 24 hours a day, and includes the following: an ultraviolet spectrophotometric analysis of 5,5' disubstituted barbiturates, chlordiazepoxide, methaqualone, diazepam, and glutethimide; qualitative urine spot tests for imipramine and phenothiazines; a blood alcohol determination. In most instances, thin-layer chromatography of urine was subsequently performed. Specific identifications of barbiturates and alcohols were performed as indicated. The routine spectrophotometric diazepam analysis performed at the time the patient was in the emergency room was as follows: a 3-ml volume of serum was buffered topH 7.4 with 0.5 ml 0.5 M phosphate buffer, and extracted with 30 ml chloroform. A 25-ml aliquot of the chloroform was washed successively with 0.45 M NaOH (discarded or used for barbiturate analysis) and 0.5 M HCl (discarded or saved for chlordiazepoxide assay). The chloroform was evaporated, the residue was dissolved in 5 ml of hexane, and the diazepam (and nordiazepam) was extracted into 4 ml 2 M HCl. The aqueous acid phase was scanned in an ultraviolet recording spectrophotometer, "diazepam" identified by its spectrum and the concentration determined by comparing the absorbance difference between 280 and 300 nm with that of similarly processed serum standards. This procedure has a coefficient of variation of 4% and is linear over a range of 1 to 20 ^ig/ml. It does not differentiate among diazepam, nordiazepam, and oxazepam, nor is it intended to be sufficiently sensitive for therapeutic monitoring. However, chlordiazepoxide, and other bases with higher/?K a s that are generally present at much lower concentrations, are removed from the chloroform with the earlier HCl wash and do not interfere. Methaqualone and its metabolites will interfere but can be recognized by their spectra. The concentrations obtained by this method, in routine use at the time of admission, were used for clinical correlation.
sufficient serum remained after spectrophotometric analysis, it was frozen for future gas chromatographic assay. In order to obtain a larger sampling for the analytic comparison, 74 additional specimens from diazepam overdose cases, from patients seen after July 1975, were also reassayed by gas chromatography but not used in the clinical study. Nordiazepam concentrations following clorazepate overdose were all determined by gas chromatography. Gas chromatography was performed as follows: 0.5 ml of plasma was alkalinized with 1 ml saturated Na 3 P0 4 8 and extracted with 8 ml hexane containing 5% acetone (V/V) and 2.5 /xg medazepam as an internal standard. After centrifugation, the solvent was transferred, and back-extracted with 1 ml 0.1 M H 2 S0 4 . After centrifugation the aqueous phase was separated, alkalinized with saturated Na 2 C0 3 , and re-extracted with 5 ml hexane:acetone (without internal standard). The solvent was separated, evaporated under a stream of nitrogen, the residue dissolved in 20 fi\ methanol, and 2 - 4 fi\ submitted to chromatography. Concentrations were determined by comparison of relative peak areas with those of extracted plasma standards. Plasma standards supplemented with exact concentrations of diazepam and nordiazepam were carried through the entire procedure.
DIAZEPAM CONCENTRATIONS IN OVERDOSE
Vol. 72 • No. 4
573
Table I. Diazepam and Ethanol Ingestion
Case
300 100-150 60 40-200
9 300 500 30 75 30 300 35 200 •) 50 200 100 7 150 90 60 1,000 7 65 50-75 7 150 80
Other Drugs
Coma Grade
Ethanol Concentration (mg/dl)
Total Benzodiazepine Concentration (/xg/ml)
Caffeine; ergotamine — — — Chlordiazepoxide (8 /ig/ml) — — — 1,000 mg amitriptyline — — Amitriptyline: perphenazine — — — — 50 mg doxepin Propoxyphene — — — — — Diphenylhydantoin (25 /xg/ml) Flurazepam — — Secobarbital (0.4 mg/dl) — —
0 0 0 0 A 0 A A A 0 0 0 0 0 0 A 0 A 0 III A 0 0 0 0 A 0 A IV 0 0
175 325 150 150 181 225 190 180 78 148 143 273 80 110 198 205 159 160 109 130 216 53 170 160 204 192 220 215 70 128 104
2 22 4 3 2.5 2.5 4 3 11 3 5 2 2 8 2 7 4 2 5 4 6 3 2 4 6 2 2 2 3 3 3
93 cases. The total benzodiazepine concentrations in this group ranged from 1 to 9 /u,g/ml. Estimated doses (by history) ranged as high as 600 mg. Sixteen patients were awake, and 22 were in grade 0 coma. The mean drug concentration of those who were awake (3.7 /Ag/ml) did not differ from that of those classified as in grade 0 coma (3.7 /u.g/ml). None of the patients who had only ingested diazepam needed medical hospitalization, regardless of drug level or doses; all were either released or transferred to psychiatric facilities after a period of observation in the emergency room.
grade 0 (asleep but arousable) and A (awake) probably reflected tolerance rather than toxicity.
Diazepam with Ethanol
Grade IV Coma
Concurrent ethanol ingestion was established by chemical analysis in 31 cases, and was the only other drug ingested in 21 of them. Table 1 summarizes the data from the 30 cases in which ethanol concentrations were greater than 50 mg/dl. Twenty-nine of the 31 patients in this group were either awake (9) or in grade 0 coma. These included four patients with diazepam levels of 6 /xg/ml or more in association with ethanol concentrations greater than 200 mg/dl. Since many of these patients were chronic drinkers, and were receiving diazepam chronically, the difference between
Seven patients were classified as in grade IV coma. All needed assisted ventilation. Total serum benzodiazepine concentrations in this group ranged from 3 to 7 £tg/ml. All of the patients in grade IV coma had ingested other drugs in addition to diazepam. In fact, as shown in Table 2, six of the seven had ingested two or more additional drugs.
Multi-drug
Ingestion
Fifty-six (61%) of the patients had ingested other drugs with diazepam. In 35 instances, additional drugs other than ethanol were also taken. Most common were short acting-barbiturates (nine cases), tricyclic antidepressants (seven cases) and phenothiazines (six cases). A drug classifiable in one of these three groups was implicated in every case with grade IV coma.
Analytic
Considerations
Comparison of the benzodiazepine concentrations determined by scanning ultraviolet spectrophotometry
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 5, 2016
5 6 10 14 15 18 19 20 21 23 26 30 31 32 36 47 49 52 53 63 67 71 76 78 79 83 87 91 92 93 95
Purported Diazepam Dose (mg)
A.J.C.P. • October 1979
JATLOW, DOBULAR, AND BAILEY
574
B.
LJ O
CD CC
° 0.8 h < 1.0 1.2 .4 200
J_ _1_ 250 300 WAVELENGTH (nm)
350
4 8 TIME (min)
12
Fio. \. A, ultraviolet absorption spectrum from serum of a patient following diazepam overdose. B, gas chromatogram from serum of the same patient, a = medazepam (internal standard); b = diazepam; c = nordiazepam. with the sum of diazepam and nordiazepam measured by gas chromatography yielded a correlation coefficient of 0.87. The mean total benzodiazepine concentrations of the 101 samples determined by ultraviolet spectrophotometry and gas chromatography were 4.97 jttg/ml and 4.70 ^g/ml, respectively. The means did not differ significantly (t test for paired observations). The ultraviolet spectrum and gas chromatographic pattern from a patient's serum are shown in Figure 1.
Parent Drug.Metabolite
Ratios
These were determined for 101 emergency-room specimens by gas chromatography, as previously described. In almost all instances the concentration of parent drug was considerably higher than that of the N-desmethyl metabolite. The ratios of parent drug to metabolite averaged 3.0. The data for those 27 patients for whom specific parent drug and metabolite concentrations were determined and clinical histories reviewed are summarized in Table 3.
Table 2. Diazepam Overdose with Grade IV Coma
Case
Diazepam Dose (mg)
16 17 25 46 30 55 92
400 ? 500 450 9 75 9
Other Drugs
Coma Grade
Total Benzodiazepine Concentration (/i.g/ml)
Imipramine, prochlorperazine propoxyphene, theophylline, ephedrine Codeine, ethchlorvynol, lithium (9.5 mg/1) 1,200 mg amitriptyline, 250 mg chlorpromazine Diphenylhydantoin, amitriptyline, propoxyphene Secobarbital (1.3 mg/dl) EthanOl (200 mg/dl), 120 mg perphenazine Secobarbital (0.4 mg/dl), ethanol (70 mg/dl)
IV IV IV IV IV IV IV
7 4 5 6 5 5 3
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 5, 2016
z 0.6
Ethanol, 180 mg/dl Barbiturate, 0.5 mg/dl; imipramine, prochlorperazine; propoxyphene Codeine, ethchlorvynol Ethanol. 180 mg/dl Ethanol, 78 mg/dl; umitriptyline Ethanol, 148 mg/dl; amitriptyline, codeine
400
150-300 500 7 500 300
1,200 mg amitriptyline; 250 mg chlorpromazine Ethanol, 110 mg/dl
60-600 30 60
2-3 g glutethimide
250 50-120
Ethanol, 205 mg/dl Secobarbital, 1.3 mg/dl Glutethimide, 0.4 mg/dl 120 mg tranylcypromine
200-250 7 7 250-500 600 7 7
Pseudoephedrine, prochlorperazine 250-500 mg chlordiazepoxide
500
Data from five cases of clorazepate overdose are shown in Table 4. These patients, none of whom required hospitalization, has serum nordiazepam concentrations ranging from 0.7 to 5.1 /xg/ml. Discussion The immediate management of possible drug overdose is, of course, supportive, with a few exceptions not specific, and relatively independent of drug identification or quantitation. An accurate drug analysis is valuable for confirmation, decisions relating to disposition, further diagnostic work-up of coma, establishing prognosis, and long-term follow-up. When confronted with a probable drug overdose, the emergency room physician will nonetheless often request a stat drug analysis, regardless of the availa-
Coma Grade
Diazepam (HSlml)
Nordiazepam (/ig/ml)
I'otal Drug (jig/ml)
0 0 0 0 0 A
3.1 1.1 0.9 1.6 8.4 0.3
0.1 0.0 0.4 0.7 0.3 1.7
3.2 1.1 1.3 2.3 8.7 2.0
IV IV A A 0 0 IV 0 0 0 A 0 0 A IV A A A A 0 A
6.2 1.0 2.8 8.5 7.8 3.9 2.7 7.5 6.6 2.6 2.6 4.4 3.1 4.2 5.0 1.8 6.0 4.2 5.4 2.8 5.0
0.6 1.7 1.1 1.2 0.0 0.2 0.2 2.4 0.8 1.7 0.3 0.4 0.1 0.2 0.9 0.3 0.6 0.4 0.6 0.5 0.6
6.8 2.7 3.9 9.7 7.8 4.1 2.9 9.9 7.4 4.3 2.9 4.8 3.2 4.6 5.9 2.1 6.6 4.6 6.0 3.3 5.6
bility of information necessary for meaninful interpretation of these data. Since diazepam reaches levels greater than 1 - 2 ptg/ml following overdose, it can under such conditions be rapidly measured by such relatively uncomplex technics as ultraviolet spectrophotometry, fluorimetry, or even gas chromatography with conventional (flame ionization) detectors. As a consequence, drug concentrations are more commonly available following diazepam overdose than, for example, in the more serious instance of tricyclic antidepressant intoxication. Since the described spectrophotometric analysis of diazepam is available at all times (24 hours/day) for cases of suspected overdose, these data were available and used for clinical correlation studies on the 93 cases. Because we wished to validate the data for use in
Table 4. Nordiazepam Concentrations Following Clorazepate Overdose Case la 2a 3 4
Purported Dose (mg) 23 tablets (? mg) 22.5 4-6 tablets (? mg) 50-300
Other Drugs Ethanol, 125 mg/dl Barbiturate (amobarbital and secobarbital), I mg/dl; ethanol, 43 mg/dl; codeine, amitriptyline Ethanol, 253 mg/dl
Coma Grade
Nordiazepam (/xg/ml)
Total Drug (/ug/ml)
A A A
5.1 0.7 2.0
5.1 0.7 2.0
II A
1.0 2.0
1.0 2.0
Diazepam (eg/ml)
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 5, 2016
17 20 21 23 24 25 32 39 40 41 43 45 47 50 51 77 80 82 85 86
Purported Dose (mg)
576
JATLOW, DOBULAR, AND BAILEY
yses was to validate the spectrophotometric data used for clinical correlation, so as to establish its suitability for emergency toxicology. The good correlation between the spectrophotometric values and the sum of diazepam and nordiazepam as measured by gas chromatography indicates that the ultraviolet spectrophotometric procedure provides a sufficiently accurate measurement of total active compound for clinical evaluation of suspected overdose. While oxazepam, the major urinary metabolite of diazepam, is also pharmacologically active, it does not reach significant serum concentrations. Spectrophotometric procedures can thus be safely recommended as a valid means to confirm that a diazepam "overdose" (excess ingestion) has occurred. It is not nearly sensitive enough for use in the therapeutic range, nor does it differentiate between diazepam and its major active metabolites. The gas chromatographic analyses were also performed to establish the relative concentrations of diazepam and its active N-desmethyl metabolite following overdose. Following chronic therapy, concentrations of nordiazepam are usually comparable to or exceed that of the parent compound. In this overdose series, concentrations of parent drug were generally well in excess of the metabolite, with an average parent to metabolite ratio of 3:1. While not surprising, this information may be useful in determining whether a high drug concentration, and perhaps mild clinical toxicity, is a result of chronic high-dose therapy or an acute overdose. Following chlorazepate overdose, only its major active derivative nordiazepam was measured. Although the number of cases is small, the data suggest that the consequences of clorazepate overdose may be no more serious than those of diazepam ingestion. These data lead to the question whether drug measurements are of any value in managing possible diazepam overdose. Experienced clinical impressions 10 and recent clinical reviews of benzodiazepine overdose 9 1 8 (without drug levels) have strongly suggested that benzodiazepine overdose is relatively benign. Nonetheless, various tabulations 120 and available manufacturer's product descriptions have indicated that serious if not fatal coma is a consequence of diazepam intoxication. Our studies indicate that diazepam overdose, confirmed by high serum drug concentrations, is not likely to produce deep coma (greater than grade I) or to necessitate hospital admission, regardless of the dose or drug level. However, lesser manifestations of central nervous system depression, such as somnolence, which also must be evaluated by the emergency room physician, can be attributed to an elevated diazepam concentration. Furthermore, despite the lack of alarming symptoms, correct diagnosis
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 5, 2016
clinical correlation studies and also to assess the relative contributions of parent drug and metabolite, a separate gas chromatographic analysis was performed. Since only 27 specimens from the original clinical correlation study were available for reanalysis, specimens from an additional 74 cases were also analyzed by both ultraviolet spectrophotometry and gas chromatography with a nitrogen detector. A retrospective clinical review for classification of central nervous system depression was not performed in this latter group of 74 cases. The conclusions to be drawn from this study are almost self-explanatory. Large overdoses with diazepam and high blood drug concentrations are generally not life-threatening, nor are they likely to produce coma. None of the patients who had ingested only diazepam were in deeper than grade 0 coma, despite serum drug concentrations well in excess of those seen following therapy. Following chronic doses of 30 mg/day, total benzodiazepine concentrations slightly more than 1 fig/ ml have been reported, 7 although daily doses of 10 mg/ day produce concentrations of less than 1 /ug/ml.7,15 The therapeutic plasma concentrations requested for adequate antianxiety effect are also reported to be < 1 jig/ml. 613 Massive chronic daily doses of 150-200 mg in one subject were reported to yield total benzodiazepine concentrations of 5-6 fx.g/m\.7 Published tables of drug levels have listed 5-20 /ug/ml as toxic and more than 20 /ng/ml as lethal. 1,20 Many of the patients in this series were in the so-called toxic range, yet did not manifest sufficient central nervous system depression to need hospitalization. Ethanol, despite its known positive interaction with many central nervous system depressants, and its reported enhancement of diazepam absorption, 12 appeared to have little impact on the sfeVerity of diazepam overdose. Concentrations of ethanol of more than 200 mg/dl in association with diazepam concentrations well above the therapeutic range were not associated with toxicity. Many of these patients did have histories of alcohol abuse, and may have had increased tolerance for central nervous system depressants. Mixed ingestions (other than ethanol) were most commonly associated with barbiturates, pheriothiazines, and tricyclic antidepressants, all of which are recognized to cause serious intoxication. Indeed, all of those patients in grade IV coma had ingested one or more of these drugs. Only case 92 is somewhat difficult to reconcile with the ingestion history, and laboratory data. In the others the clinical manifestations (Table 2) could reasonably be attributed to the other additional drug(s), regardless of the diazepam level. One reason for the large-scale comparison between the spectrophotometric and gas chromatographic anal-
A.J.C.P. • October 1979
Vol. 72 • No. 4
DIAZEPAM CONCENTRATIONS IN OVERDOSE
2. 3. 4. 5. 6.
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 5, 2016
centrations of diazepam and its metabolite N desmethyldiazepam in relation to anxiolytic effect. Clin Pharmacol Ther 15:473-483, 1974 7. DeSilva JA, Koechlin BA, Bader B: Blood level distribution patterns of diazepam and its major metabolite in man. J Pharm Sci 55:692-702, 1977 8. DeSilva JAF, Bekersky I, Puglisi CV, et al: Determination of 1,4-benzodiazepines and diazepine-2-ones in blood by electron capture gas chromatography. Anal Chem 48:10-19. 1976 9. Greenblatt DJ, Allen MD. Noel BJ, et al: Acute overdose with benzodiazepine derivatives. Clin Pharmacol Ther 21:497-514. 1977 10. Greenblatt DJ, Shader RI: Psychotropic drug overdose. Manual of Psychiatric Therapeutics. Edited by RI Shader. Boston. Little. Brown, 1975, pp 237-268 11. Greenblatt DJ, Shader RI: Benzodiazepines in Clinical Practice. New York Raven Press, 1974 12. Hayes SL, Pablo G, Radomski T. et al: Ethanol and oral diazepam absorption. N Engl J Med 296:186-189, 1977 13. Hillestad L, Hansen T, Melsom H, et al: Diazepam metabolism in normal man. I. Serum concentrations and clinical effects after intravenous, intramuscular and oral administration. Clin Pharmacol Ther 16:479-487. 1974 14. Jatlow P: UV spectrophotometry for sedative drugs frequently involved in overdose emergencies. Methodology for Analytical Toxicology. Edited by I Sunshine. Cleveland. Ohio Chem Rubber Co., 1974. pp 414-420 15. Kaplan SA, Jack ML. Alexander K, et al: Pharmacokinetic profile of diazepam in man following single intravenous and oral and chronic oral administrations. J Pharm Sci 62:1789-1976. 1973 16. Reed CE, Driggs MF. Foote CC: Acute barbiturate intoxication: References A study of 300 cases based on a physiologic system of clasBaselt RC, Wright JA, Cravey RH: Therapeutic and toxic consification of the severity of the intoxication. Ann Intern Med centrations of more than 100 toxicologically significant drugs 37:290-303, 1952 in blood, plasma, or serum: A tabulation. Clin Chem 21:44-62, 17. Regent TA, Wahl KC: Diazepam abuse: Incidence, rapid screen1975 ing and confirming methods. Clin Chem 22:889-891. 1976 Bell EF: The use of naloxone in the treatment of diazepam 18. Roche Scientific Report: Incidence of fatalities associated with poisoning. J Pediatr 87:803-804, 1975 Valium (diazepam) alone or in combination with other subBenzodiazepine overdose. Med Lett Drugs Ther 18:60, 1976 stances. Hoffman-LaRoche Inc, Nutley, N. J.. 1976 Berger R, Green B, Mclnick A: Cardiac arrest caused by oral 19. Sparks H, Goldman AS: Diazepam intoxication in a child. Am J diazepam intoxication. Clin Pediatr 14:842-844, 1977 Dis Child 109:128-129. 1965 Cate JC, Jatlow PI: Chlordiazepoxide overdose. Interpretation 20. Winek CL: Tabulation of therapeutic, toxics and lethal concenof serum drug concentrations. Clin Toxicol 6:553-561, 1973 trations of drugs and chemicals in blood. Clin Chem 22:832— Dasberg HH, van der Kleign E, Guelen PJR, et al: Plasma con836. 1976
of diazepam overdose is, of course, important for longterm follow-up, including psychotherapy, particularly in prevention of a second attempt with, perhaps, a more potent compound. Following our earlier study of another benzodiazepine, chlordiazepoxide (Librium), we reached similar conclusions about overdose with that drug."' In summary, whether a diagnosis of diazepam overdose is based upon history or drug measurement, or both, when coma more severe than grade I is present, another drug or explanation should be sought. There will, in all probability, be exceptions to this generalization among susceptible individuals (e.g., patients with myxedema, those with chronic pulmonary disease, the very elderly and the very young). Serious diazepam overdose with coma has been reported to occur in small children. 2 4 1 9 Drug measurements are still useful in establishing that an acute ingestion has occurred, and to explain mild central nervous system depression. Despite the suggested utility of diazepam measurements, our data suggest that there is rarely a justifiable need for an ultrastat or desperate emergency serum diazepam analysis. 1.
577
