Monday 13 July 2009

July's Case 2009- The Answers

Dear Bloggers

I hope that you found this case thought provoking.

Professor Masami Matsumura has kindly considered this month's case despite his busy schedule. I thank him for his time and input.

His answers to the blog case are as follows:

Thank you for showing difficult case again. I would try to solve.

The patient was a 64-year-old lady.
Problem list on admission
#1Benzodiazepine overdose
Diazepam 1,000mg and Lorazepam 200 mg
Altered mental status, GCS 3/15
Tone decreased throughout
Reflexes generally depressed
No response to pain
#2Aspiration pneumonia
Fever
Tachycardia
High RR [respiratory rate]
Hypoxia
#3History of anxiety and insomnia

Problem list on second episode
#1 Unresponsive
#2 Shock
#3 Hypoxia
#4 Acidemia, respiratory acidosis
#5 Tachycardia
#6 After flumazenil administration

Question 1: Why did the patient lapse into unconscious despite initial improvement on the evening of admission to the hospital?

Benzodiazepine overdose will last for several days and flumazenil is a short-acting drug. I think second episode was caused by benzodiazepine overdose again.

Question 2: Why did the patient deteriorate so quickly during resuscitation, with profound hypotension and sinus tachycardia >200/min and why did it resolve?

PSVT with high rate can cause hypotension. Additional flumazenil administration and intubation can alleviate patient condition better.

Question 3: What acid-base disturbance is demonstrated by the blood gas?

Hypoxia and respiratory acidosis with hypercapnea caused by benzodiazepine overdose.

Question 4: What additional therapies may have helped this patient when admitted to the hospital that could have been commenced in the ER department?

Activated carbon administration can absorb overdose of benzodiazepine in GI tract.


What had happened in this case? (Question 1-3 are covered in the following text)


Firstly, we must look at the drugs. This patient has both long and short acting benzodiazepines. Remember that benzodiazepines get metabolised into other active benzodiazepines that can have longer half-lives. Hence, these drugs can have sedative effects for a prolonged period of time. The effects are more marked in the elderly who have reduced ability to excrete these drugs. For example, diazepam has a half-life of 20-40 hours but its active metabolite is Nordazepam which has a half-life of 60 hours!! The overall duration of action is about 24-48 hours !

Nordazepam is converted to Oxazepam (the same compound produced from Temazepam) which is then combined with glucuronic acid and excreted renaly. On the other hand, Lorazepam is a shorter acting benzodiazepine although is half life is 8-12 hours. However, it has no active metabolite so its overall half is the same.

The action of Flumazenil is rapid and hence it is no surprise that the admitting resident considered the patient 'stable' after injecting the 'antidote'. However, Flumazenil only acts for a short period of time (maximum of 2 hours) and the actions of the long acting benzodiazepines can return. This is indeed what happened here.

The benzodiazepines caused unconsciousness and respiratory compromise despite the respiratory rate of 24 per minute. The patient had decreased air entry on examination and hypopnea despite tachypnoea. This resulted in CO2 retention and acute respiratory acidosis.

However, if one looks at the ABG result, the HCO3 level is minimally elevated and base excess is -5. These suggest an added metabolic acidosis. This is unsurprising in the presence of severe hypoxaemia. Her calculated bicarbonate should have been 36.2 if a pure respiratory acidosis were present.

The history, physical exam and vital signs suggested that she had SIRS (systemic inflammatory response syndrome) as part of a septic syndrome. The fact that her blood pressure dropped on minimal incline suggested hypovolaemia and / or septic shock.

The patient having vomited and aspirated was evidence enough of an unsafe airway and the patient should have been intubated on admission prior to administration of Flumazenil. Moreover, such patients need to have activated charcoal. Most texts suggest that the activated charcoal be given within 1 hour of an overdose. This is not entirely accurate because some drugs have a slower absorption e.g. Lorazepam, and the use of charcoal can have benefits even after 1 hour. Moreover, in view of the unconscious state, the charcoal could only have been administered with the patient being intubated to protect the airway.


Why did this patient drop her blood pressure and develop a rapid tachycardia?

The side effects of Flumazenil include hypotension and dysrrhythmia aside from the other serious side effect of seizures in patients taking benzodiazepines chronically. In essence, too much Flumazenil had been administered.

Typical guidance suggests starting with 0.2mg then waiting 30 seconds for a response. Following that give 0.3mg and wait another 30 seconds for a response. Following that, 0.5mg can be administered.

Patients may sometimes require Flumazenil every 20 minutes in view of the short half life of the drug especially when long acting benzodiazepines have been taken in overdose. However, this depends entirely on how the patient responds to treatment. Hence, in view of the short lived hypotension and tachycardia, it is likely that Flumazenil being given at too high a dose was to blame here.


Question 4: What additional therapies may have helped this patient when admitted to the hospital that could have been commenced in the ER department?
  • Patients should receive activated charcoal even after 1 hour post-overdose and further doses can be administered several hours later if considered appropriate.
  • If the patient has a GCS of less than 8/15, intubation should be initiated to protect the airway and also to allow for activated charcoal to be administered. Remember though, an ET tube is not completely 100% protective against aspiration. Also remember the saying -- 'if the GCS is less than 8 then intubate'
  • Flumazenil should be used with extreme caution because of its adverse side effects and should be given in low dose and titrated up. Despite the ability to reverse the effects of the benzodiazepines, its actions are short lived and repeated doses may be necessary. Remember though, benzodiazepines do not usually cause significant respiratory depression making them relatively safe in overdose. Most patients just sleep off the effects of this class of drug. It is when respiratory depression occurs then the antidote should be considered and especially when taken in combination with other CNS depressants.
  • Remember that Flumazenil is CONTRAINDICATED in mixed overdoses with Tricyclic Antidepressants as the effects of the latter drug can be increased.
  • Patients with sepsis often require large amounts of fluid. It is appropriate to administer adequate amounts of fluid and for BP and Pulse to be checked regularly in addition to urine output to assess volume status. In reality, in patient with sepsis, it is better to place a central venous line to assess volume requirements rather than guessing. A CVP of 8-12 cmH20 would be considered an appropriate goal to achieve in sepsis.
Take Home Message

The moral of the story here is that we should not be complacent that this is just 'another overdose patient'. If we do not take precautions and manage such patients appropriately, then problems can and do occur to the detriment of the patient.

In cases of overdose, the patient should be managed as a minimum in a high-intensity emergency medicine admissions unit with experienced staff or on an HCU or ITU . Full monitoring equipment should be utilised. Patients should not been placed on a general ward in a side room.

Physicians need to take more care to find out the mental state of the patient before prescribing medications that can be used in overdose and they should not be given out like candy. A thorough drug history is required to be taken by the doctor and drug-drug interactions need to be considered before adding in another medication.

Drug-drug interactions can be easily checked with PDA software e.g. ePocrates

Many thanks to Prof Matsumura.