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

Continuous plasmafiltration in sepsis syndrome. Plasmafiltration in Sepsis Study Group.

Reeves JH, Butt WW, Shann F, et al.

Crit Care Med. 1999;27:2096-104. [abstract]

Reviewed by Steven Cray, MBBS FRCA, Birmingham Children's Hospital, UK

Review posted February 4, 2000

---

I. What is being studied?:

The study objective:

To investigate in patients with sepsis the effect of plasmafiltration on survival, number of organs failing and the humoral inflammatory response.

The study design:

Multicenter randomized controlled trial.

The patients included:

Thirty patients (22 adults, 8 children) with a clinical diagnosis of sepsis syndrome for < 24 hrs at one of 7 intensive care units. Sepsis syndrome was diagnosed according to accepted criteria. (*)

The patients excluded:

Patients were excluded if they had positive HIV serology, uncomplicated meningococcal meningitis in children (because of a generally good prognosis) or had undergone cardiopulmonary bypass in the last 48 hrs.

The interventions compared:

Patients in the plasmafiltration group received continuous plasmafiltration for 34 hrs with a total volume of plasma exchange of 250 ml/kg. Plasma filtrate was replaced with a mixture of fresh frozen plasma, albumin and electrolyte solution. Patients in control and plasmafiltration groups also received standard intensive care treatment. Conventional treatment followed guidelines agreed upon by consensus of participating hospitals.

The outcomes evaluated:

The primary outcome measure was 14 day survival. In addition the authors evaluated the number of organs failing after 7 days using accepted dichotomous criteria and measured the concentration of a number of inflammatory mediators in the blood at 0, 6, 24 and 48 hrs from study entry. In the plasmafiltration group, filtrate was collected after 3 hrs of filtration to estimate sieving coefficients of these mediators.

II. Are the results of the study valid?

Primary questions:

1. Was the assignment of patients to treatments randomized?

Yes. A block method of randomisation was used designed to allocate patients equally to the plasmafiltration or control groups according to investigative center and the presence of immune deficiency.

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

Was followup complete?

Yes.

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

Yes. The authors state that the data were analysed on an intention-to-treat basis. There does not seem to have been any crossover of patients between groups.

Secondary questions:

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

No method of blinding was used. The authors make no comment about this, but it would certainly be a challenge to perform effective blinding of a treatment such as plasmafiltration.

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

APACHE II scores (PRISM in children) in both groups were similar at trial entry, as was the calculated risk of death. Because of the small number of patients included, statistical tests produced non-significant differences between the groups for other factors, for example, 6/16 control patients had refractory shock compared with 3/14 plasmafiltration patients.

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

Intensive care management of both groups was according to an agreed protocol. This included the management of cardiovascular support, ventilation, antibiotics, nutrition and renal replacement therapy. The authors do not state the extent of non-compliance with this protocol. Because the personnel caring for the patients were unblinded, we cannot be sure that both groups were treated equally.

III. What were the results?

1. How large was the treatment effect?

Eight of 14 patients (57%) in the plasmafiltration group and 8 of 16 controls (50%) survived for 14 days (P=0.73). This corresponds with a point estimate for absolute reduction in the risk of death in the plasmafiltration group of 7% and for relative risk reduction of 14%.

There was no difference in the mean number of organs failing in the first 7 days (2.57 in plasmafiltration group vs. 2.94 in controls P=0.37).

Plasmafiltration did not influence mean concentrations of interleukin-6, granulocyte colony-stimulating factor, thromboxane B2, total white cell count, neutrophil count, or platelet count, but it was associated with significant reductions of alpha-1-antitrypsin, haptoglobin, C-reactive protein, and complement fragment C3 in the first 6 hrs (p < 0.05). The sieving coefficients (concentration of solute in filtrate divided by its concentration in plasma) for all inflammatory mediators approached one.

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

This is an underpowered study and thus the confidence intervals for the size of the treatment effect are wide. Calculation of the 95% confidence intervals for absolute risk reduction with plasmafiltration produces a range from a 43% increased chance of survival at 14 days to a 29% increased risk of death. In reality, we can only conclude that this study was too small to exclude the possibility of either substantial benefit or substantial harm from the use of plasmafiltration in sepsis.

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

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

Not directly. The study included some children, but mainly involved adult patients.

2. Were all clinically important outcomes considered?

Recent large trials in sepsis have tended to use 28-day mortality as the primary endpoint. In fact the authors of the present study appear to have continued to follow their patients progress beyond the 14 day cut-off and a Caplan-Meier survival curve is provided. The estimated 28-day survival is 42% in the plasmafiltration group and 40% in the control group.

It would have been interesting for the authors to have measured the concentrations of a number of other mediators which are believed to be important in sepsis (e.g., interleukin-1, tumor necrosis factor-alpha, platelet activating factor)

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

The present study does not enable firm conclusions to be drawn about the efficacy of plasmafiltration in sepsis syndrome because of the small numbers of patients recruited. Although the authors report a negative result, there are insufficient patients to reliably detect even a 50% reduction in the relative risk of death.

Plasmafiltration exposes the patient to a number of risks from vascular access, anticoagulation, mechanical failure of the extracorporeal circuit and transmission of infection by blood components. The potential costs of fresh frozen plasma and albumin are substantial - indeed the study was brought to a premature halt because of the inability of the Australian blood transfusion service to supply enough albumin. In addition, the cost of pumps, filters and staff training must be considered.

A number of therapies designed to alter the systemic inflammatory response in sepsis have been investigated in large randomized controlled trials, without evidence of increased survival. [1] It may be that this approach is fundamentally wrong, or that targeting these therapies in certain groups of patients might show benefit. On a simplistic level, it is an attractive proposition to remove plasma containing inflammatory mediators and replace this with "clean" plasma. However, it is interesting to note from the present study that plasmafiltration did not alter the blood concentration of a number of mediators (e.g., IL-6), despite a high sieving coefficient.

There is some (rather weak) evidence from case series that plasmafiltration may lead to improved survival in children with meningococcemia [2] and there are some centers that employ plasmafiltration commonly in this group of patients. Although plasmafiltration is a practical proposition in children, in a retrospective review of the experience from Melbourne [3], its use did not lead to improved hemodynamic stability or biochemistry. There is currently no evidence from randomized controlled trials to support the use of plasmafiltration in children as a standard intensive care treatment

The authors of the present study propose to undertake a multicenter randomized controlled trial of plasmafiltration in children with septic shock. The estimate of the sample size for this trial to show a 10% reduction in mortality with plasmafiltration is 900.

References

1. Abraham E. Why immunomodulatory therapies have not worked in sepsis. Intensive Care Med 1999; 25: 556-566. [citation]
2. van Deuren M, Santman FW, van Dalen R, Sauerwein RW, Span LF, van der Meer JW. Plasma and whole blood exchange in meningococcal sepsis. Clin Infect Dis 1992;15:424-30. [abstract]
3. Reeves JH, Butt WW. Blood filtration in children with severe sepsis: safe adjunctive therapy. Intensive Care Med 1995;21:500-4. [abstract]

Footnote

Diagnosis of sepsis syndrome in adults required all of the following: 1) clinical evidence of infection, 2) respiratory rate >20 breaths/min or minute volume >10 L/min, 3) heart rate >90 beats/min, 4) temperature >38 C or < 35.5 C and 5) one or more of: a) altered mental state, b) PaO2/FiO2 < 280, c) lactic acidemia, d) oliguria < 0.5 ml/kg/hr. In children the critical value for heart rate was adjusted to: 1-2 yrs, 155 beats/min; 2-3 yrs, 140 beats/min; 3-7 yrs, 120 beats/min; 7-10 yrs, 110 beats/min; 10-12 yrs, 100 beats/min; >12 yrs, 90 beats/min. The respiratory criteria were replaced by "clinical requirement for mechanical ventilation". [return to text]

Comments

Back to the EB Journal Club Index