It’s one of those time honored treatments that most hospital-based providers are familiar with. The banana bag, reserved for intoxicated patients presenting to the ED or admitted to the hospital. They’ve been around so long, we just take them for granted. But like most things that have become dogmatic, they are due to be questioned from time to time.
A banana bag is a proprietary mix of “good” stuff, including electrolytes and vitamins, especially thiamine and magnesium. The exact content varies from hospital to hospital. Thiamine and other B vitamins give the resulting solution the characteristic color, hence the term “banana.”
Does it actually do good things like ward off Wernicke’s encephalopathy and megaloblastic anemia? A paper from Jacobi Medical Center in the Bronx prospectively evaluated a series of intoxicated people entering their ED. They drew vitamin B12, folate, and thiamine levels to see if they were deficient enough to even need vitamin supplementation.
Here are the factoids:
- These folks (only 77 patients) were very drunk! Average BAC was 280mg/dL.
- Vitamin B12 and folate levels were not critically low in any patient
- Thiamine was low in 15% of patients, but none had clinical evidence of a deficiency
- Later review of prior visits revealed that some patients with low levels had received a previous banana bag within 1 month. Did it do any good?
Bottom line: Most of our intoxicated patients are not vitamin deficient, and don’t need supplementation. The real kicker is that we almost never really try to find out if the patient might be a chronic abuser and potentially at risk. We just hang the bag. Remember, everything we do in medicine has a potential downside. And if the patient really doesn’t need a banana bag in the first place, there is no benefit to balance that risk. The next time you ask for that little yellow bag, think again!
Reference: Vitamin deficiencies in acutely intoxicated patients in the ED. Am J Emerg Med 26(7):729-795, 2008.
All US trauma centers verified by the American College of Surgeons are required to have programs for identifying patients who may have alcohol problems and for providing brief intervention (BI) and referral to therapy. Typically they use a standard interview tool (or the fact that patient blood alcohol exceeded a certain limit) to determine if brief intervention is indicated. If so, a trained professional (social worker, nurse, psychiatrist) sits down with the patient for a counseling session that may last 30 minutes, give or take. The idea is that the intervention has more impact in the face of the recent traumatic event, and the patient will be less likely to offend again. A number of studies have shown that alcohol consumption and risk-taking behavior decrease, at least in the short term, for patients who are taken to an emergency department and receive BI.
But does brief intervention really work for people who have been arrested for driving under the influence (DUI) but not injured? Researchers at UC Davis looked at 200 first-time arrestees for DUI in a county jail during a 1 month period. They randomized them for BI or no BI, and 181 of the 200 enrollees actually finished a 90-day followup, which is very good. AUDIT was used to measure the degree of problematic drinking (scale 0-40, higher means worse).
Here are the interesting factoids:
- Mean blood alcohol was 0.14 mg/dl, which is a bit on the low side
- Average initial AUDIT score was about 8 in both control and brief intervention groups
- AUDIT score decreased by 3.4 in controls and 4.7 in BI subjects (not significantly different)
- The likelihood of binge drinking, abstinence, alcohol-related injury, and seeking treatment was no different between the groups at 90 days.
Bottom line: Adding a brief intervention session to the routine after someone has been jailed for DUI does not appear to work. Although the study numbers are small, the number needed to show a difference appears to be pretty large, so the result is probably real. What this means is that jail does change behavior in first-time offenders, and brief intervention doesn’t add that much. I’ve always marveled at the fact that we try to modify behavior with just one counseling session. Much of the substance abuse literature indicates that ongoing counseling and support is needed for real problem users, and patients with alcohol related injuries don’t appear to be an exception.
What exactly is the CIWA protocol?
It is a tool used commonly in the US that helps clinicians assess and treat potential alcohol withdrawal. A significant amount of injury in this country is due to the overuse of alcohol. A subset of these patients are admitted and do not have access to alcohol. They may begin to withdraw within a few days, and this condition can lead to dangerous complications.
The Clinical Institute Withdrawal Assessment measures 10 items that are association withdrawal:
- Nausea / vomiting
- Paroxysmal sweats
- Tactile disturbances (itching, bugs crawling on skin, etc)
- Visual disturbances
- Auditory disturbances
All items are measured on a scale of 0-7 with the exception of orientation, which uses a scale of 0-4. All subscores are tallied to arrive at the final score.
The total score is used to determine whether benzodiazepines should given to ameliorate symptoms or avoid seizures. Typically, a threshold is selected (8 or 10) and no medications are needed as long as the patient is under it. Once it is exceeded, graduated doses of lorazepam or diazepam are given and vital signs and CIWA scores are repeated regularly. The protocol is discontinued once the patient has three determinations that are under the threshold.
The individual dosing scale and monitoring routine varies by hospital. Look at your hospital policy manual to get specifics for your institution.
For a copy of the CIWA scoring criteria, click here.