Harm Statin therapy is associated with fewer deaths in patients with bacteraemia.
This review may be edited
Summary
Posted By:	Gordon S Doig
E-Mail:	Gordon.Doig@EvidenceBased.net
Posted Date:	7 Nov 2007
Title:	Statin therapy is associated with fewer deaths in patients with bacteraemia.
Authors:	Kruger P, Fitzsimmons K, Cook D, Jones M, Nimmo G.
Reference:	Intensive Care Med. 2006 Jan;32(1):75-9.
Link:	Click here for a direct link to the paper. A password may be required for access to fulltext.
Abstract:	OBJECTIVE: Beneficial effects with statin use are increasingly reported in a variety of patient groups. There is in vitro and clinical evidence for its antiinflammatory and immunomodulatory therapeutic roles. We aimed to assess the association between statin administration and mortality in bacteraemic patients. DESIGN: A retrospective cohort analysis. SETTING: A 300-bed acute general hospital. PATIENTS AND PARTICIPANTS: All patients (n=438) requiring hospital care for an episode of bacteraemia during the years 2000-2003 were included. Statin use, patient outcome, and clinical and laboratory variables were collected. INTERVENTIONS: None. MEASUREMENTS AND RESULTS: There was a significant reduction in all-cause hospital mortality (10.6% vs. 23.1%, p=0.022) and death attributable to bacteraemia (6.1% vs. 18.3%, p=0.014) in patients who were receiving statin therapy at the time of bacteraemia (n=66). The reduction in all-cause hospital mortality (1.8% vs. 23.1%, p=0.0002) and death attributable to bacteraemia (1.8% vs. 18.3%, p=0.0018) was more pronounced in the patients who continued to receive statin therapy after the diagnosis of bacteraemia (n=56). The apparent mortality benefit persisted after controlling for differences between the groups. Statin use prior to admission was associated with a reduced adjusted hospital mortality rate (odds ratio 0.39; CI 95% 0.17, 0.91; p=0.029), and continuing statin use after bacteraemia increased this effect (odds ratio 0.06; CI 95% 0.01, 0.44; p=0.0056). CONCLUSION: This retrospective study demonstrates a significant survival benefit associated with continuing statin therapy in bacteraemic patients. The potential for statins as an adjuvant therapy in sepsis warrants further investigation.
Are the Results Valid?
1. Were there clearly identified comparison groups that were similar with respect to important determinants of outcome, other than the one of interest?
	The authors identified patients with positive blood cultures through the micro dept culture results. Table 1 presents a reasonable list of 'baseline variables' but because this study is retrospective, they are restricted to what could be found in the hospital charts. Physiological parameters such as heart rate, resp rate, temp and WBC are reported in the on-line supplementary material available here.
2. Were the outcomes and exposures measured in the same way in the groups being compared?
	Yes. Use of statins was obtained from the charts and the primary outcome, mortality, was obtained in the same way in both groups.
3. Was follow-up sufficiently long and complete?
	Yes, day 28 mortality is a reasonable outcome.
4. Is the temporal relationship correct?
	No. The authors report identifying 66 patients who were receiving statins at the time of bacteremia however the primary multivariate analysis is conducted on only 56 statin patients. 10 patients were excluded from the primay analysis because they had their statins stopped during their hospital stay. Six (60%) of these 10 patients died. If the primary purpose of this study is to determine the presence of a causal relationship between statin use and outcome, it is important to investigate these 10 patients further. If these patients had their statins d/c because they became too sick to take oral meds, then statin use may simply be a marker for the ability to take oral meds. Thus, it is possible the mortality rate observed in the remaining patients who did not have their statins d/c (1/56) is simply an indicator that they always remained well enough to take oral meds.
5. Is there a dose response gradient?
	No, this was not investigated.
What are the Results?
1. How strong is the association between exposure and outcome?
	The authors report a significant reduction in mortality in patients who received statins and did not become sick enough to have them d/c (OR = 0.058). The removal of the sub-group of patients who had statins d/c introduces a significant potential for bias. At the very least, the appropriate control group would be other patients who became sick enough to require oral meds d/c.
2. How precise is the estimate of risk?
	The 95% CI ranges from an OR of 0.0008 to 0.43
Will the Results Help Me In Caring For My Patients?
1. Are the results applicable to my practice?
	Caution should be exercised due to the potential for bias resulting from removing patients who had statins d/c. Six of these ten patients died.
2. What is the magnitude of the risk?
	An OR of 0.058 is very large. If the baseline mortality is 20 %, an OR of 0.058 results in an NNT of 5.8.
3. Should I attempt to stop the exposure?
	If a patient is too sick to tolerate oral intake, statins cannot be delivered.
What other people had to say about: Statin therapy is associated with fewer deaths in patients with bacteraemia. Add Your Comment
[ Back ] Any questions or comments please contact Gordon S. Doig Implemented and designed by John Soer and Gordon Doig Page last modified on Friday August 10, 2001 www.EvidenceBased.net
	THIS IS A SPACE MESSAGE. THIS IS STILL THE BEST METHOD TO SPACE A TABLE FOR NETSCAPE AND INTERNET EXPLORER. IF YOU SEE THIS MESSAGE CONTACT WEBADMIN@EvidenceBased.net