Ghid de urgență pentru sindromul de sevraj alcoolic. De la evaluare la tratament: scale clinice și protocolul cu lorazepam

1. Legend – clinical flow of the scales

Part I: Hierarchical algorithm for clinical scale assessment

Stage 1. At admission: PAWSS

Prediction of Alcohol Withdrawal Severity Scale – 10 binary items (Yes/No) evaluating history, biology and clinical findings at admission. The only prospectively validated instrument for predicting complicated withdrawal (sensitivity 93.1%, specificity 99.5%).

• Score < 4: low risk → standard monitoring.
• Score ≥ 4: high risk → aggressive prophylaxis (BZD front-loading, parenteral thiamine, monitoring every 1-2 hours).

Stage 2. In the emergency department: BAWS

Brief Alcohol Withdrawal Scale – five items (tremor, sweating, anxiety, agitation, hallucinations/orientation), scored 0-3. Rapid triage in under 1 minute, correlation > 0.9 with CIWA-Ar.

• Score < 5: reassessment at 2 hours.
• Score ≥ 5: immediate BZD initiation; transfer to a specialized ward.

STAGE 3a. On the ward: CIWA-Ar (Sellers Inventory)

The gold standard for severity assessment – 10 items, maximum score 67. It requires a cooperative patient.

• 0-8: mild withdrawal → monitoring.
• 9-15: moderate → BZD at score ≥ 10.
• > 15: severe → increased risk of complications.

Stage 3b. Uncooperative patient/cardiovascular risk: Cushman

Seven items with vegetative predominance (heart rate, blood pressure, respiratory rate, tremor, sweating, agitation, sensorium). Independent of patient cooperation.

• < 7: mild.
• 7‑14: moderate → treatment.
• > 14: severe → major risk.

Stage 4. Continuously: RASS

Richmond Agitation-Sedation Scale – 10 steps (+4 to −5). The safety umbrella for the entire protocol.

• Target: 0 to −1 (alert/mildly drowsy).
• RASS ≤ −2: STOP benzodiazepines regardless of CIWA-Ar/Cushman score – risk of respiratory depression.

2. Synthetic table — thresholds and actions

From assessment to treatment

Once severity and risk have been hierarchically established through the clinical scales, the pharmacological decision becomes concrete. In Romanian clinical practice, where oxazepam is not pharmaceutically available, lorazepam (Anxiar®) is the only benzodiazepine with an optimal pharmacokinetic profile for patients with hepatic dysfunction, the elderly, or those with respiratory insufficiency. Part II of this guide translates the assessment algorithm into a concrete therapeutic protocol, calibrated on the Cushman score and under the safety umbrella of the RASS scale⁽¹²⁾.

Part II: Lorazepam therapeutic protocol

3. Context and rationale

In Romanian clinical practice, oxazepam is not pharmaceutically available, and lorazepam (Anxiar®) remains the only alternative with pure conjugative hepatic metabolism and no active metabolites. This pharmacokinetic characteristic makes it the benzodiazepine of choice for three essential patient categories^(1,2,13):

• Patients with hepatic cirrhosis or hepatic failure (Child-Pugh B and C).
• Elderly patients (> 60 years old) – physiological decline of hepatic oxidative metabolism.
• Patients with respiratory insufficiency (COPD, aspiration pneumonia, OSAS) – reduced risk of accumulation.

Unlike diazepam and chlordiazepoxide (oxidized through the cytochrome P450 system at the microsomal level, with accumulation of active metabolites such as desmethyldiazepam, with a half-life of 30-100 hours), lorazepam is directly conjugated with glucuronic acid and eliminated renally in an inactive form (half-life 10-20 hours)⁽³⁾.

4. Pharmacological equivalence diazepam ↔ lorazepam

The equivalence ratio recognized internationally in alcohol withdrawal protocols is:

5 mg diazepam ≈ 1 mg lorazepam (oral).

This ratio is the one adopted in protocols from the University of California, San Francisco, Vanderbilt University Medical Center, and UCH-Denver, as well as in the criteria defining benzodiazepine-refractory withdrawal (>10 mg lorazepam equivalent in 1 hour or >40 mg lorazepam equivalent in 4 hours)^(4,5,9).

I.v. route rule: the oral dose is halved for parenteral administration (bioavailability nearly 100% i.v. versus approximately 90% oral, but the faster plasma peak requires lower initial doses)⁽⁶⁾.

5. Oral lorazepam protocol based on the Cushman score

5.1. Moderate withdrawal (Cushman score 7-14)

Note: Unlike diazepam, lorazepam does not undergo metabolic auto-tapering, which is why the taper must be performed explicitly, reducing both the unit dose and the dosing frequency.

5.2. Severe withdrawal (Cushman score >14)

Maximum dose: 20 mg lorazepam/day (equivalent to 100 mg diazepam).

Exceeding this threshold should prompt consideration of benzodiazepine-refractory withdrawal and the introduction of adjunctive agents (phenobarbital, dexmedetomidine, propofol in ICU)^(5,7,10).

6. I.v. lorazepam protocol (uncooperative/severely agitated patient)

Indications: disturbances of consciousness, severe agitation precluding oral administration, lack of compliance, severe Cushman score. Unlike diazepam, lorazepam can be administered via:

• slow i.v. bolus (1 minute per vial);
• i.m. (an advantage over diazepam, which has erratic i.m. absorption due to its lipid solubility);
• sublingual (s.l.) – useful when venous access is difficult.

6.1. Standard administration

6.2. Mandatory safety measures

• RASS monitoring every 15 minutes after each i.v. dose; the patient must remain arousable at all times.
• Antidote ready: flumazenil 0.3 mg i.v. bolus over 15 seconds, repeatable every 60 seconds up to a maximum of 2 mg.
• Resuscitation kit at hand (Guedel airway, Ambu bag, oxygen source, pulse oximeter).
• Beware of propylene glycol: i.v. lorazepam preparations contain propylene glycol as a solvent, which at high doses (>10 mg/24 h) can produce metabolic acidosis, renal failure or hyperosmolarity^(8,11). Monitor serum osmolality and osmolar gap at cumulative doses >10 mg/day i.v.
• In case of delirium: combine with tiapride or haloperidol (2-5 mg).

7. Clinical pitfalls specific to lorazepam

a) Shorter half-life = more frequent reassessment

Lorazepam has a half-life of 10-20 hours, compared with diazepam (half-life 30-100 hours, including active metabolites). Reassessment with the Cushman/CIWA-Ar scale must be performed every 2-4 hours, NOT every 6-8 hours.

b) Breakthrough symptoms

Due to the absence of metabolic auto-tapering, episodes of breakthrough withdrawal may occur between fixed doses. The solution is symptom-triggered titration rather than rigid fixed schedules.

c) Reduced risk of accumulation in the elderly

The absence of active metabolites makes lorazepam significantly safer in elderly patients. Nevertheless, increased cerebral sensitivity to BZDs requires reducing the initial dose by 25-50% (0.5-1 mg at first administration).

d) Predictable i.m. absorption

Lorazepam has rapid and complete i.m. absorption, unlike diazepam – a significant clinical advantage when venous access is difficult.

e) Beware of rebound

Abrupt discontinuation of lorazepam may produce more pronounced anxious rebound than diazepam. Gradual tapering is mandatory.

f) Propylene glycol in i.v. preparations

Accumulation of propylene glycol at high cumulative doses (>10 mg/day i.v. for >24 h) is a rarely discussed iatrogenic pitfall⁽¹¹⁾. In ICU patients with refractory withdrawal, this issue justifies early transition to phenobarbital⁽¹⁰⁾.

8. Critical memento for on-call duty – integrated synthesis

The cards below synthesize the essential rules that link the assessment (Part I) to the actual treatment (Part II). They are designed to be printed and laminated at the on-call room bulletin board.

PAWSS ≥ 4 = “red flag”

Frequent visual monitoring; anticipated preparation of emergency medication; initiation of prophylaxis even before the appearance of severe symptoms.

B1 BEFORE glucose

Always 500 mg thiamine i.m./i.v./day before any glucose infusion. No exceptions, regardless of the score on clinical scales.

The 5 → 1 Rule

5 mg diazepam = 1 mg lorazepam orally. For i.v.: halve the dose (5 mg diazepam = 0.5 mg lorazepam i.v.). Maximum dose: 20 mg lorazepam/24 h.

Reassessment every 2-4 hours

The short half-life of lorazepam requires frequent monitoring with Cushman/CIWA-Ar. Diazepam tolerates 6-8 hours; lorazepam does NOT.

SLOW i.v. means SLOW

Lorazepam i.v.: minimum 2 minutes for 1-2 mg. Rapid administration produces apnea and hypotension. Flumazenil ready (0.3 mg i.v. bolus).

RASS ≤ −2 = STOP

Regardless of the Cushman/CIWA-Ar score, reaching RASS −2 requires stopping benzodiazepines. Risk of respiratory depression.

The sleep rule

DO NOT wake the patient to administer BZDs. If the patient is asleep, the CNS is sufficiently inhibited. Reassess 30-60 minutes after spontaneous awakening.

Propylene glycol – caution in ICU

At cumulative i.v. doses > 10 mg/day of lorazepam, monitor serum osmolality. Risk of metabolic acidosis and renal failure.

Advantage in cirrhosis and the elderly

Lorazepam: direct conjugation, no active metabolites, no cytochrome P450. Indicated in Child-Pugh B/C, elderly > 60 years old, COPD, OSAS.

Conflict of interests: none declared

Financial support: none declared

This work is permanently accessible online free of charge and published under the CC-BY.