Abstract
Benzodiazepines (BZD), typically used to treat insomnia and anxiety disorders, are largely prescribed in all medical fields. This widespread use has led to misuse through frequent and often inappropriate consumption and iatrogenic overdosing.
The risk of physical dependence, highly related to the duration of treatment and pharmacokinetic profile of BZD, such as its short half-life, varies between 15% and 44% of chronic users (Ashton, 2005). Symptoms related to BZD discontinuation syndrome can be severe and in most cases may preclude the patients stopping the drug. Seizure represents the most severe, although rare, complication of BZD withdrawal (Chouinard, 2004; Marriott and Tyrer, 1993).
Although BZD intentional abusers very often have other substance abuse problems (O’brien, 2005), the rising level of addiction from legitimate prescription is largely underestimated. Our opinion origins from the increasing number of patients seeking an inpatient detoxification treatment in our Addiction Unit (up to approximately 50% of clinical admissions). Our target patients are BZD chronic abusers, and applied treatment is based on rapid stopping of BZD under slow flumazenil infusion (FLU-I) for 8–10 days (Gerra et al., 2002).
We read with great interest the paper of Hood et al. published in your journal (Hood et al., 2009). The detoxification treatment described in the article is similar to Gerra et al.’s and Quaglio et al.’s (Gerra et al. 2002; Quaglio et al. 2005).
We fully agree with Hood et al. that FLU-I appears to be a highly effective treatment resulting in withdrawal discomfort of lesser severity than any other treatment cessation currently available. We also agree that the treatment needs further development before becoming a routine treatment of BZD dependence but, in our opinion, FLU-I is not so safe as suggested by Hood et al. (2009). We would like to stress the seizure risk related to a detoxification protocol, underestimated by both Gerra et al. and Hood et al. During the last 7 years, our group treated 286 chronic users (156 males and 130 females, median age of 42) of high doses of BZD with FLU-I. The median daily dose was 14 times greater than the maximum dose allowed. Congruent with Gerra et al. and Hood et al., patients immediately stopped abusing BDZ (lormetazepam in 54% and lorazepam in 21% of all cases) and rapidly tapered low doses of other BZD (generally clonazepam) in 3 days up to no BZD on day 4. FLU-I (at a rate of 1–2 mg/24 h) started on day 1 and lasted up to discharge at day 8–11. We can confirm that symptoms were reported as mild, requests to stop FLU-I were infrequent and only lasted for very few hours; no patients withdrew from the treatment.
The high risk of seizure remains the only unresolved problem related to this highly effective treatment. In our experience, we observed seizures in 10 patients, 3.5% of all cases (6 out 10 withdrawn from lormetazepam, 2 from lorazepam, 1 from alprazolam and 1 from zolpidem): 5 patients suffered of tonic-clonic crises (in 1 case repeated) and 5 patients presented with generalized epileptic crises characterized by absence with mild motor signs. All but tonic-clonic cases required a transfer in neurologic unit where symptoms lasted for several days. Seizures happened from day 4 to day 10 of FLU-I. Only 4 out of 10 were dependent from other substances (2 out 4 were heavy drinkers, 1 used heroin and 1 misused butalbital) and only 1 patient had an episode of seizure in his anamnesis (withdrawn from lorazepam).
In conclusion, FLU-I can be an effective and rapid treatment of BZD overdosing dependence. In our opinion, seizure risk is high and remains an unmet issue. Further studies are needed to prevent this serious complication (perhaps using anticonvulsant prophylaxis or different doses of FLU-I), but, at the moment, we strongly recommend to not use FLU-I in outpatient settings as suggested by Gerra et al. and, partially, by Hood et al.