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

Plasmapheresis in severe sepsis and septic shock: a prospective, randomised, controlled trial

Busund R, Koukline V, Utrobin U, Nedashkovsky E.

Intensive Care Med 2002; 28: 1434-143 [abstract]

Reviewed by Jambunathan Krishnan, Vinay Vaidya, University of Maryland, Baltimore

Review posted December 24, 2002

---

I. What is being studied?:

The study objective:

To determine the therapeutic efficacy and safety of plasmapheresis in the treatment of patients with severe sepsis and septic shock.

The study design:

Prospective randomized, clinical trial with a planned, mid-study, interim analysis at the City Emergency Hospital #1, Archangels, Russia.

The patients included:

Patients in the ages of 17 to 70 years with severe sepsis or septic shock. Sepsis was diagnosed using the "Bone" criteria(1). Severe sepsis was defined as sepsis associated with organ dysfunction, Hypoperfusion abnormality, or sepsis induced hypotension (systolic BP < 90 mmHg). Septic shock was defined as sepsis induced hypotension, persisting despite adequate fluid resuscitation, along with the presence of hypoperfusion abnormalities or organ dysfunction. Patients with hypoperfusion abnormalities on inotropes were also included in the septic shock category even if they had normal blood pressures. A total of 106 patients were included in the study.

As an aside, the consent process was also intriguing - they obtained delayed consent in those unconscious patients rather than get consent from next of kin. It is rather difficult to consent or say no to randomization and plasmapheresis after the fact. Deferred consent is becoming increasingly difficult to accept, at least in the United States.

The patients excluded:

Patients were excluded if:

1. they had been treated for severe sepsis or septic shock in other hospitals for > 12hrs before being transferred to the study hospital
2. they had severe underlying disease. Severe underlying disease included terminal cancer, terminal cardiac failure, end stage renal failure and potentially lethal injuries.

The interventions compared:

The interventions being compared here are

1. Conventional treatment for sepsis and
2. Plasmapheresis in addition to conventional treatment for sepsis

All patients in both groups received the conventional treatment for sepsis according to the indication in each case .This included antibiotics, fluid resuscitation, surgical procedures, and cardiovascular and ventilatory support as indicated. The study group received plasmapheresis in addition to the conventional treatment.

The outcomes evaluated:

The primary outcomes being studied were 28 day survival and safety of plasmapheresis. Patients were followed for 28 days or until they died. The secondary outcome was adverse events reflecting safety.

II. Are the results of the study valid?

Primary questions:

1. Was the assignment of patients to treatments randomized?

Yes. The patients included in the study were block randomized in two stages to allow an interim analysis after inclusion of 50 patients. However, the authors do not describe the exact method by which randomization was achieved. e.g., whether a random number table or computer generated numbers were used.

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

Was followup complete?

Yes, all 106 patients in the study were accounted for and appear to have been adequately followed up.

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

Yes, there were no crossovers and the patients were analyzed in the groups they to which they were randomized.

Secondary questions:

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

The procedure of plasmapheresis is such that it makes it difficult if not impossible for the patients and health workers to be blinded. The authors do not mention if the study personnel involved in statistical analysis were blinded to the treatment.

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

The two groups were similar in all aspects except for the following variables;

1. Age: the plasmapheresis group was younger (mean age 41 +/- 15 vs. 48 +/- 16 in the control group, p value = 0.03),
2. Site of infection: plasmapheresis group had more abdominal infections (33 patients in the plasmapheresis group had abdominal infections versus 16 in the control group, p value = 0.04)
3. Mechanical ventilation: in the plasmapheresis group fewer patients required mechanical ventilation; 46% vs. 67% p value = 0.03) (See table1 and table 2, page 1436). This would imply the plasmapheresis group was not as ill however, APACHE III scores were similar.

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

Yes, apart from the plasmapheresis (and use of fresh frozen plasma), it appears that the two groups were treated equally as far as the use of surgery, inotropes and anticoagulation (see Table 2, page 1436). However, few real details are provided other than raw numbers and broad categories. There is no discussion of fluid or nutritional management, and there is no evidence provided that similar strategies were followed with antibiotics, inotropes, diuretics, steroids or mechanical ventilation.

Every patient in the plasmapheresis group obviously received fresh frozen plasma, while only 43 of 82 patients in the control group were given fresh frozen plasma (p value = 0.001). The only other difference was the use of heparin in doses of 200 to 300 Units/kg to keep activated clotting time between 250 and 300 seconds during the duration of plasmapheresis (approximately 2 hours) Although over 90% of patients in both groups received heparin

III. What were the results?

1. How large was the treatment effect?

The 28-day all-cause mortality was 33.3% (18 of 54 patients died) in the plasmapheresis group versus 53.8% (28 of 52 patients died) in the control group resulting in an absolute risk reduction of 20.5% (95% CI = 2% to 39%) This equals a relative risk (RR) of mortality in the plasmapheresis group of 0.61 (95% CI = 0.39 to 0.97). The number of patients needed to be treated with plasmapheresis to prevent one death was 4.9 (95% CI 2.5 to 50). On correcting for variables that were significantly different between the groups at baseline, the effect of the treatment variable on mortality lost classical statistical significance with an odds ratio of 0.41 with a 95% CI of 0.15-1.09 and a p value of 0.07

A post hoc sub-group analysis performed on the subgroup of patients with abdominal infections showed a significant reduction in mortality with an absolute risk reduction of 35.2% (95% CI 6%, 62%) ..

The second outcome studied was the safety of plasmapheresis. There were no fatal adverse reactions attributable to the plasmapheresis procedure, but 17% of this group had an adverse event. Six patients had short periods of hypotension during the procedure and one patient had an allergic reaction to Fresh frozen plasma. Two patients in the plasmapheresis group died of bleeding 4 and 17 days after the procedure.

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

As mentioned earlier the RR of mortality in the plasmapheresis group was 0.61 (95% CI = 0.39 to 0.97) Based on the data available the relative risk reduction for fatal outcome in the plasmapheresis group was 39% with the calculated 95% CI ranging from 3% to 61%. The number needed to treat is 4.9 with a 95% CI ranging from 2.5 to 50. After logistic regression to control for significant differences between the groups, the 95% confidence intervals for the odds ratio were rather wide and crossed unity, from 0.15 to 1.09.

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

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

The study population appears to be similar to that seen in adult ICU's in most academic centers and the results can be applied to an adult population. However these results cannot be transposed to a pediatric population. This intervention needs to be studied in a pediatric population separately before we can apply the results in that segment. However this study does give us significant ideas for further clinical research in children.

2. Were all clinically important outcomes considered?

Yes, the clinically important outcomes were considered in this study. Survival is good outcome to study in this case as it is an objective criteria and therapy studies need good strong objective criteria. The adverse reactions of the treatment offered were also watched for and the authors did not find any significant adverse outcome. There was no comment about neurologic complications: could some of these patients have had small intracranial hemorrhages or strokes?

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

This study offers an interesting treatment option in the management of sepsis. The Number Needed to Treat (NNT) as per this study is 4.9 (95% CI 2.5, 50) a significantly low number, but the fact that the potential advantage of the therapy disappeared after correction for differences in the study and control group precludes its utility. Even if we disregard the correction for the differences in the two groups the upper limit of the 95% CI of 50 raises doubts in one's mind about the utility of the intervention. However it may be an option worth thinking about and studying further. And although described an "undramatic," 11% of the plasmapheresis group had hypotension, making this a potential harm that needs to be explored further in this group of labile patients.

References

1. Bone RC, Balk RA, Cerra FB, et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. Chest 1992;101:1644-1655. [abstract]

Comments

Back to the EB Journal Club Index