Criteria abstracted from The Users' Guide to Medical Literature, from the Health Information Research Unit and Clinical Epidemiology and Biostatistics, McMaster University

Highlighted lines and questions below provide links to the pertinent description of criteria in The EBM User's Guide, now available at the Canadian Centres for Health Evidence

Article Reviewed:

Please visit the new Evidence Based Journal Club Reviews

High-Dose Antithrombin III in Severe Sepsis: A Randomized Controlled Trial.

Warren BL, Eid A, Singer P, et al.

JAMA 2001;286:1869-1878. [abstract]

Reviewed by Kiran Hebbar MD, Tony Petrillo MD, Children's Healthcare of Atlanta at Egleston Children's Hospital, Emory University, Atlanta, GA.

Review posted September 2, 2004

---

I. What is being studied?:

The study objective:

To determine if high dose Anti-thrombin III (given within 6 hours of onset) would improve survival in patients with severe sepsis and septic shock.

The study design:

This was an extremely well designed study. It was a randomized, double blind, placebo-controlled, phase 3 clinical trial, conducted at 211 centers in 19 countries, between March 1997 and January 2000, which enrolled 2314 patients with severe sepsis (KyberSept Trial).

The patients included:

Adult men and women greater than 18 years of age whom:

1. Gave informed consent with a 6 hour period
2. Clinical evidence of sepsis with a suspected source of infection
3. Body temperature (rectal or core) higher than 38.5 C or lower than 35.5 C
4. Leukocyte count higher than 10 X 10³/microliter or lower than 3.5 X 10³/microliter
5. In addition three of the following six signs had to be met within the six hour enrollment period
   1. Tachycardia - heart rate grater than 100
   2. Tachypnea - respiratory rate greater than 24 or intubation
   3. Hypotension - Systolic blood pressure (SBP) lower than 90 despite sufficient fluid replacement or the need of vasoactive agents to maintain SBP greater than or equal to 90.
   4. Thrombocytopenia - platelets less than 100 X 10³ /microliter
   5. Elevated lactate levels - above upper limit of normal or metabolic acidosis (pH less than 7.3 or base excess - 10mmol/L) in the absence of respiratory alkalosis.
   6. Oliguria - urine output less than 20 mL/hour despite fluid resuscitation.

The patients excluded:

This is an extensive list (Warren p. 1871). The important points to note are the factors that may contribute to a bleeding complication. Despite their exclusion even the use of low dose heparin was associated with higher number of bleeding events.

1. Treatment with an antithrombin III concentrate within the last 48 hours
2. Treatment with heparin (except subcutaneous low dose or intravenous line flushing) or coumadin derivatives
3. Non-steroidal anti-inflammatory drug treatment within the previous 2 days
4. Known bleeding disorder or ongoing massive surgical bleeding
5. Platelet count less than 30 X 10³/microliters

The interventions compared:

Comparison was made between the use of Antithrombin III (Aventis Behring, Margurg, Germany) and placebo (1% human serum albumin) of equivalent volume. Patients were randomly assigned to receive 30,000 IU antithrombin III total, with an initial loading dose of 6000 IU followed by continuous infusion of 6000 IU per day for 4 days.

The outcomes evaluated:

The primary evaluated outcome was death 28 days after commencement of therapy. Secondary outcome measures included survival time, length of intensive care unit stay, and occurrence of new organ dysfunction (according to Logistic Organ Dysfunction score) within 7 days. Surgical interventions and bleeding events were noted for the first 28 days. Bleeding was classified as major if it was intracranial or at least 3 units of blood were required.

II. Are the results of the study valid?

Primary questions:

1. Was the assignment of patients to treatments randomized?

Yes. Randomization was done in a 1:1 manner, through a central randomization center available 24 hours a day. Randomization plans had been prepared in advance in blocks of 4 patients.

2. Were all patients who entered the trial properly accounted for and attributed at its conclusion?

Was followup complete?

Yes. The authors are able to account for all patients considered/enrolled in the study. Of the 2341 eligible patients, 2 were excluded because they met exclusion criteria. Of the remaining 2339 patients, 1166 were assigned to receive placebo, out of which 9 patients were withdrawn for either not receiving placebo (8) or withdrawing consent (1). 1173 patients were assigned to receive Antithrombin III, of which 16 patients were withdrawn for not receiving the medication (13), withdrawing consent (1), or 28 day survival unknown (2). The primary efficacy population consisted of 2314 patients assigned into equal numbers of 1157 to either receive study drug or receive placebo.

Were patients analyzed in the groups to which they were randomized?

Yes, however of the 1157 patients included in the primary efficacy analysis to receive placebo, there were 261 protocol violations: 82 completed < 90% of the study treatment, 115 received prohibited concomitant medication (high dose non-steroidal anti-inflammatory, high dose heparin, or additional Antithrombin III), and 100 violated inclusion/exclusion criteria. Patients included in the primary efficacy analysis to receive antithrombin III had 287 protocol violations; 91 completed < 90% of the study treatment, 100 received prohibited concomitant medication, and 105 violated inclusion/exclusion criteria Patients treated according to protocol was considered a subgroup of interest to the authors.

Secondary questions:

3. Were patients, health workers, and study personnel "blind" to treatment?

Yes. Patients, health workers, and study personnel were blinded to treatment assignment. Investigators contacted the randomization center and were told the medication package number to be used. Packages for individual patients consisted of vials and labels identical in appearance for antithrombin III and placebo.

4. Were the groups similar at the start of the trial?

Yes. There was no significant difference between the 2 groups for age, sex, simplified acute physiology score II (SAPS II), body weight, and race. Both groups were well matched for source of infection, frequency of bacteremia, and type of infecting microorganism.

5. Aside from the experimental intervention, were the groups treated equally?

Unknown. There was no standardized therapeutic protocol for other aspects of critical care management, including fluid management, inotrope or pressor use, antibiotic therapy or ventilator strategy. The authors do not report any difference in management or the number of patients enrolled by various centers.

III. What were the results?

1. How large was the treatment effect?

None. The antithrombin III group mortality rate did not differ from the placebo group in any SAP II risk stratum or in the overall study population (38.9% for antithrombin III vs. 38.7% for placebo; p = 0.94). Kaplan Meier plots of survival over the 90-day study period did not show any significant differences between the two treatment groups. However, a trend toward a reduction in 90-day mortality in the antithrombin III group was seen in the high risk SAP II stratum (predicted mortality 30-60%).

Analysis of prespecified subgroups showed no significant difference in the relative survival benefit between the antithrombin III and placebo groups after 28 days from study entry. Prespecified groups included : 1) patients treated according to protocol, 2) severity of sepsis (moderate, high, or very high), 3) patients with baseline antithrombin III < 60%, and 4) concomitant heparin use.

Comparison between treatment groups for the incidence of new organ dysfunction as described by Logistic Organ Dysfunction score showed that new cardiovascular dysfunction within 7 days of study entry was significantly more likely to occur in the antithrombin III group than the placebo group in the primary efficacy population. One proposed explanation by the investigators is the association with bleeding events (hemorrhage), which occurred in 34.5% of antithrombin III patients with new cardiovascular dysfunction and 11.9% of placebo patients with new cardiovascular dysfunction.

Analysis of the subgroup with no concomitant heparin showed the 90 day mortality rate for the antithrombin III patients without heparin was 44.9% compared to 52.5% for the placebo patients without heparin.

Mortality in patients who received concomitant heparin was looked at separately and showed patients who received low dose heparin did not respond as favorably to antithrombin III.

2. How precise was the estimate of the treatment effect?

These estimates were quite precise. The 95% confidence interval for relative risk reduction for 28-day mortality in the primary efficacy population was 0.91-1.11% suggesting no benefit for those treated with high dose antithrombin III.

IV. Will the results help me in caring for my patients?

1. Can the results be applied to my patient care?

No. This study was confined to patients aged 18 and older. More importantly, data from the study did not show any benefit to the use of antithrombin III. The higher incidence or bleeding and new onset cardiovascular complications in patients receiving antithrombin III would deter clinicians from utilizing the drug. The only subgroup that had a trend toward benefit were those with severe SAPS II scores.

The 28-day mortality in patients who did not receive concomitant heparin was 37.8% of the antithrombin III group and 43.6% of the placebo group. This along with the trend toward less renal and pulmonary dysfunction in the antithrombin III group should be more closely examined in future studies.

This phase III clinical trial shows high antithrombin III failed to demonstrate significant benefit despite very promising phase II data providing evidence of its efficacy in sepsis.

2. Were all clinically important outcomes considered?

No. Although the primary outcome analyzed was appropriate (mortality), other important outcome variables, which may be critical in determining the cost benefit analysis of this drug, were not included. These include duration of, pressor requirement and ventilator days. Length of ICU stay was a secondary outcome. No difference was found between groups.

3. Are the likely treatment benefits worth the potential harms and costs?

No. The total percentage of patients with adverse events and serious adverse events was high (46%). With exception of the bleeding events these adverse effects did not differ between these two groups. Bleeding complications occurred in 22% of the antithrombin III patients and 12.8% of the placebo group. The difference was most marked in the groups receiving heparin. Post-op surgical patients in both groups were more likely to develop bleeding complications within the first 7 days.

When analyzed by effect on mortality, it is clear that antithrombin III is not beneficial and should be used in patients. It is impossible to assign a dollar based cost benefit analysis with current information.

Comments

Back to the EB Journal Club Index