The severe sepsis bundles as processes of care: a meta-analysis

Objective: The use of the sepsis bundles in patients with severe sepsis and septic shock has been controversial in the last decade. Clinical studies have reported beneficial, as well as negative results. We conducted a meta-analysis to assess the clinical evidence and to evaluate survival effects.

Data source: Database searches (2004-current) of Medline, CINAHL, Pubmed, Cochrane, Scopus and Google scholar databases which covered full publications, abstracts from conferences and digital thesis were performed using the search terms sepsis, septic shock and/or bundles, processes of care, guidelines, early goal directed therapy, resuscitation.

Results: From 253 identified studies, 21 sepsis bundle original studies were selected and included 23,438 patients. The Resuscitation 6 hour Bundle pooled analysis (1819 patients) achieved the greatest survival benefit (odds ratio (OR) 2.124, 95% CI 1.701-2.651, p<0.000) with the Management 24 hour Bundle pooled analysis the lowest survival benefit (16,521 patients) (OR 1.646, 95% CI 1.036-2.614, p<0.035). Both bundles together (Complete Bundle) achieved a combined survival benefit (OR 1.744, 95% CI 1.421-2.141, p<0.000). ScvO2 and blood glucose components were analysed individually to assess their contribution to survival.

Conclusion: The Resuscitation 6 hour bundle in the context of the patient population at hand is unlikely to do harm and is yet to be established in primary research in Australia. The Management 24 hour Bundle could not establish a strong enough survival benefit above current routine practice. The sepsis guidelines and bundles have demanded more credible process measurements and debate to induce positive changes in the intervention and treatment care of patients with severe sepsis.