A summary table comparing patients across the three benzodiazepine categories alprazolam, diazepam, and other benzodiazepine for each of the outcome variables is shown in Table 1. The multivariate model for the outcomes of ICU admission, mechanical ventilation, coma and flumazenil administration, produced by logistic regression is shown in Table 2. ICU admission, mechanical ventilation and flumazenil administration were significantly more likely for patients who had ingested alprazolam compared with those who had ingested other benzodiazepines after adjusting for other potential predictor variables Table 2.
However, alprazolam ingestion was not associated with an increased risk of coma compared to other benzodiazepines Table 2. The multivariate model for the outcome of LOS, produced by linear regression, is shown in Table 3. The table can be interpreted best by taking the exponential of the coefficients. Hence, LOS is predicted to be 1.
LOS increases by 1. The number of total benzodiazepine prescriptions has remained constant for the period — at approximately 8. However, the numbers of alprazolam prescriptions rose over the same period from 0. The indications for which various benzodiazepines were prescribed in general practice are presented in Table 4. Forty-eight percent of alprazolam prescriptions were for panic disorder, but this was the given indication in just 0.
The incidence of benzodiazepine overdose has mirrored their clinical popularity, increasing from their introduction in the s to peak in the s [ 27 ]. In our study the incidence of deliberate self poisoning with alprazolam rose from the late s, peaking in the mid s and then remaining constant Figure 1. This partially reflects the increasing prescription rate over the decade. The clinical effects of benzodiazepine overdose were minor in the majority of cases, but alprazolam was significantly more toxic that other benzodiazepines, based on a range of outcome measures.
Alprazolam had a longer median LOS, greater ICU admission rate, greater requirement for mechanical ventilation and more cases in which flumazenil was administered. Although a number of factors could account for the greater toxicity of alprazolam in overdose, our study provides evidence that alprazolam may be intrinsically more toxic than other benzodiazepines.
By using multivariate analysis we adjusted for biologically plausible confounders, including dose, age and co-ingestion of other sedative agents. A previous study suggested that the variable toxicity of benzodiazepines may be related to different rates of absorption [ 29 ]. This earlier study did not specifically examine alprazolam focusing on temazepam and oxazepam [ 29 ].
In our study alprazolam not only increased LOS and ICU admission rate, but also increased the use of interventions to prevent complications of respiratory depression: The relatively greater toxicity of alprazolam in overdose raises a number of questions about its increasing use almost trebled in 10 years and the population in which it is mainly prescribed. Alprazolam was approved for use in panic disorder in in the United States and made available on the Pharmaceutical Benefits Scheme in Australia for the same condition in [ 2 , 30 ].
Concerns regarding the potential over-prescription of alprazolam for panic disorder have been raised [ 30 ]. Because suicidal ideation and suicide attempts are more prevalent in people with panic disorder than in the general population [ 31 ], the use of alprazolam in this group needs to be better controlled. It has been argued that while those receiving alprazolam are at increased risk of suicidal behaviour by virtue of their psychiatric diagnoses , alprazolam actually diminishes suicidal ideation [ 1 ].
The presentation of medications in terms of quantity available and packaging is important for the risk and severity of deliberate self poisoning [ 32 ]. Initially available in 0. In Australia alprazolam DDDs per prescription is available in quantities four times as large as diazepam 25 , twice as large as nitrazepam 50 or oxazepam 50 and more than three times as large as flunitrazepam Thus a person taking alprazolam has access to relatively large quantities of easily ingestible medication.
This is of significant concern because this patient group may be at a higher risk of deliberate self poisoning and we have demonstrated that alprazolam is more toxic than other benzodiazepines. The relatively greater toxicity of alprazolam in deliberate self poisoning and its current status as among the two benzodiazepines available in largest quantity per prescription, support restrictions on pack size and limitation of availability in bottles. In sufficiently large doses benzodiazepines can cause coma, respiratory depression [ 3 ] and death [ 33 ].
Severe toxicity is often related to co-ingestants, especially alcohol and opiates, and advanced age is an additional risk factor for severe toxicity [ 3 , 29 , 33 , 34 ]. The increased toxicity of benzodiazepine overdose with co-ingestion of other sedating medications and increasing age was also seen in our study. This suggests that the relative toxicity of alprazolam compared with other benzodiazepines is not related to the level of consciousness that occurs in overdose as measured by the GCS.
This may be either due to GCS being a poor measure of coma in overdose patients or that the reason for alprazolam's increased toxicity is more dependent on respiratory depression. The predictors of flumazenil administration were different to other outcomes reflecting the use and contra-indications of this antidote. It is unclear why female sex was a predictor of its use. That flumazenil was used significantly more often in alprazolam overdose compared with other benzodiazepines is a significant finding because in the Hunter Area Toxicology Service the use of flumazenil is restricted and is generally only given by or at the request of the attending clinical toxicologist.
This policy may confound the assessment of risk factors for flumazenil use. In addition to demonstrating that alprazolam was more toxic than other benzodiazepines, our study showed that co-ingestion of other medications also significantly worsened outcome. TCA co-ingestion was by far the most important and over-shadowed the effect of alprazolam vs other benzodiazepines. This is to be expected based on the known significant toxicity of TCAs.
In addition other co-ingestants affected the outcomes, but the effect was of a similar order of magnitude to the difference between alprazolam and other benzodiazepines. There were a number of limitations of the study. Although a study of pure ingestions provides the best indicator of the intrinsic toxicity of a drug, it is less useful in providing information on the majority of overdoses where more than one drug is ingested.
The study included a large number of cases overall and used multivariate analysis to adjust for the effect of major co-ingestant drugs, so still provided a good indication of intrinsic drug toxicity. In addition, observational studies such as this are likely to have good external validity since they are conducted in the clinical context in which care is provided [ 25 ]. Although the increase in LOS is unlikely to be clinically significant, the increased ICU admission rate and use of mechanical ventilation is a significant increase in resource use.
Drug concentrations were not available for patients in the study because they are not part of the routine management of patients by the Hunter Area Toxicology Service. However, all poisoned patients admitted to the Hunter Area Toxicology Service have the drug of ingestion confirmed by history taking on at least two occasions, and this is confirmed with history from ambulance officers, family and friends, as well as evidence of empty drug containers. A study of quetiapine overdose in the Hunter Area Toxicology Service has validated this by demonstrating that the peak concentration of quetiapine following overdose highly correlated with reported dose [ 35 ].
In conclusion, our study demonstrates that alprazolam is relatively more toxic than other benzodiazepines. Because this effect remains after adjustment for dose, co-ingested medication and age, this greater toxicity appears due to intrinsic toxicity of alprazolam. Alprazolam overdose should be regarded as more significant than other benzodiazepines. In addition, the pack size and packaging of alprazolam should be reviewed to decrease the risk of it being taken for deliberate self poisoning.
Geoffrey Isbister designed the study, supervised the data extraction, contributed to data analysis, interpretation and all drafts of the paper. David Sibbritt did all statistical analysis and contributed to all drafts. Ian Whyte designed the database used in the study and contributed to all drafts of the paper. Debbie Whyte and Toni Nash did all data entry. Stuart Allen did all data extraction from the database. We would like to acknowledge Stuart Allen for extracting the data from the database, and Debbie Whyte and Toni Nash for entering the data into the database.
National Center for Biotechnology Information , U. Br J Clin Pharmacol. Received Oct 20; Accepted Dec This article has been cited by other articles in PMC. Abstract Aims To describe alprazolam poisoning and the relative toxicity of alprazolam compared with other benzodiazepines. Methods A database of consecutive poisoning admissions to a regional toxicology service was searched to identify consecutive benzodiazepine deliberate self poisonings, which were coded as alprazolam, diazepam or other benzodiazepine.
Results There were single benzodiazepine overdose admissions: Conclusions Alprazolam was significantly more toxic than other benzodiazepines. Methods Hunter Area Toxicology Service database The Hunter Area Toxicology Service is a regional toxicology unit situated at the Newcastle Mater Misericordiae Hospital that services a population of about people and is a tertiary referral centre for a further [ 23 ].
Study population The study population included all benzodiazepine overdose admissions between January and October that were a result of a deliberate self poisoning. Data analyzed From the database, the following information was obtained: Prescription data Prescription data for Australia were obtained from the Health Insurance Commission for the years — data not available prior to Results During the study period there were single benzodiazepine overdose admissions of which were deliberate self poisonings.
Alprazolam overdose There were cases of alprazolam overdose that presented to the Hunter Area Toxicology Service between January and October Number of benzodiazepine overdoses presenting each year to the Hunter Area Toxicology Service for — Comparison of alprazolam to other benzodiazepines in overdose A summary table comparing patients across the three benzodiazepine categories alprazolam, diazepam, and other benzodiazepine for each of the outcome variables is shown in Table 1.
A comparison of patients across the three drug categories alprazolam, diazepam and other benzodiazepine and each of the outcome variables. A multivariate model for predicting the natural logarithm of LOS in benzodiazepine overdose. Benzodiazepine prescription data The number of total benzodiazepine prescriptions has remained constant for the period — at approximately 8.
Indications for prescription of alprazolam and other benzodiazepines taken from The Copernicus Knowledge System recorded between 1st July and 30th September by general practitioners. Discussion The incidence of benzodiazepine overdose has mirrored their clinical popularity, increasing from their introduction in the s to peak in the s [ 27 ].
Acknowledgments Geoffrey Isbister designed the study, supervised the data extraction, contributed to data analysis, interpretation and all drafts of the paper. There were no conflicts of interest for any author. The patient's symptoms were attributed to ongoing benzodiazepine withdrawal and through titration of 2 mg doses of diazepam, he was stabilised on a regime of 10 mg per day. He was discharged home for weekly follow up by medical and nursing staff from his community mental health team.
In response to the patient's request to stop benzodiazepines, a reducing strategy was agreed. However, his compliance was erratic and 3 weeks later while unable to sleep, he inflicted serious stab wounds to his neck and chest. The patient was admitted to intensive care for 10 days and transferred to psychiatric care for an additional 3 weeks.
Two recent studies have examined the role of benzodiazepines on actual or attempted suicide. While disinhibition and paradoxical reactions inducing suicidal impulses were considered, the authors believed that confounding by the original indication was a more likely explanation. In this patient's presentation, factors regarding both the properties of benzodiazepines and monitoring his response appear relevant. Although benzodiazepine dependence syndrome was correctly diagnosed, the extent of his usage was not fully appreciated.
This is particularly important when changing from a short-acting benzodiazepine to those with a longer half-life, as the equivalent dose for 1 mg of lorazepam is 10 mg of diazepam. Similarly, the second act occurred while on a reducing dose of diazepam with poor compliance. Benzodiazepine-associated paradoxical disinhibition, an uncommon but recognised phenomenon characterised by acute excitement and an altered mental state, is an additional consideration.
While the mechanism of paradoxical disinhibition is poorly understood, the most widely accepted explanation is that they occur secondary to inhibition of the restraining influences of the cortex and frontal lobe. While the dramatic and potentially fatal consequences of this patient's acts of self harm appear to be rare in benzodiazepine withdrawal, prescribers should remain aware that dependence on benzoadiazepines can develop quickly, particularly with short-acting agents.
The case also emphasises the need for close liaison between secondary and primary services, particularly when changes in medication are advised. National Center for Biotechnology Information , U. Br J Gen Pract. Abstract Benzodiazepines are commonly prescribed in primary care for anxiety disorders and insomnia. Notes Consent The patient has consented to the publication of this report. Competing interests The authors have stated that there are none.
A communication to all doctors from the Chief Medical Officer. Schweizer E, Rickels K. Benzodiazepine dependence and withdrawal: Acta Psychiatr Scand Suppl. The benzodiazepine withdrawal syndrome. The role of benzodiazepines in elderly suicides. Scand J Public Health. Aggressive dyscontrol in patients treated with benzodiazepines.