• 
• 

Three Part Question

In [patients presenting to the emergency department in septic shock] is [norepinephrine] better than [dopamine]?

Clinical Scenario

A 47-year-old female presents to the emergency department with fever, shortness of breath, tachycardia and tachypnea. Her O2 saturation is 90% with an Fi02 of 50% and her blood pressure is 75/35 mmHg. Her x-ray reveals a large consolidation compatible with the diagnosis of community-acquired pneumonia. After adequate fluid resuscitation and early antibiotics, she remains hypotensive and poorly perfused. You have dopamine and norepinephrine at hand for vasopressor support, and you wonder which one is the best in septic shock.

Search Strategy

A. A BestBETs was found on a similar topic but the data has changed since this publication.
Robinson, A, Med J 2010;27:556e558. doi:10.1136/emj.2010.097329, http://bestbets.org/bets/bet.php?id=1813
B. There is an updated Cochrane review on the use of vasopressors in shock that was not included in the previous BestBETs
C. Embase
D. Pubmed
E.The website clinicaltrials.gov was searched for an ongoing trial on the topic. None was found
C. Embase search results:
#1 \'norepinephrine\'/exp OR norepinephrine 140,311
#2 \'dopamine\'/exp OR dopamine 305,884
#3 \'septic shock\'/exp OR \'septic shock\' 50632
#4(#1 AND #2AND #3) 874
#5 (#1 AND #2AND #3 AND \'meta analysis\'/de OR \'systematic review\') 27

D. Pubmed search results:
#1 septic shock: 24632
#2 norepinephrine: 98798
#3 dopamine: 128588
#1 and #2 and #3: 173

Search Outcome

A total of 3 publications (recent systematic reviews with meta-analysis) were found to answer our question. We limited our search to human subjects and English language.

Search conducted on December 30th, 2012, revised on January 17th, 2013

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Havel C, et al. Cochrane 2011 Austria	6 vasopressors 3212 patients in shock from all causes. 2 sub-group analyses comparing dopamine to norepinephrine were pertinent to our PICO question: 1- mortality analysis; 2- side effects analysis. mortality analysis: norepinephrine vs Dopamine 1400 patients with septic shock except De Baker with shock. Ruokonen 1993 Martin 1993 Marik 1994 Mathur 2007 Patel 2010 De Backer 2010 Side effect analysis: norepinephrine vs dopamine 1931 patients with septic shock except De Baker with shock. Patel 2010 De Backer 2010	Systematic Review that included 23 RCTs. 6 RCTs(5 RCTs in patients with septic shock and one RCT included patients with all causes of shock (De Backer 2010). 2 RCTs out of 6 RCTs (1RCT included patients in septic shock and one RCT included patients with shock at large (De Backer 2010).	12 months mortality	No difference in mortality between dopamine compared to norepinephrine RR 1.05; 95% CI [0.97 - 1.15]	Among the 6 RCTs included for this analysis, the largest study (De Backer 2010) accounting for 85.5% weight in the analysis did not only include patients in septic shock. This study included 62% of patients in septic shock, 17% of patients with cardiogenic shock, 16% of patients in hypovolemic /hemorrhagic shock and 5% in shock from other causes. This confounds the results for our question about septic shock patients. Unclear risk of bias for 5 studies identified by review authors (mostly risk of selection of bias). However, the summary result is mainly influenced by the De Backer study (n=1036 for mortality at 12 months and n=1679 for arrhythmia at 28 day) that fulfills all low bias risk criteria. 2 RCT only for arrhythmia analysis.
			Adverse effects measured at 28 days	Dopamine is associated with more arrhythmia RR 2.34; 95% CI [1.46 - 3.78]
Vasu T.S, et al 2012 USA	norepinephrine vs dopamine 2043 patients with septic shock except De Baker with shock Ruokonen 1993 Martin 1993 Marik 1994 Mathur 2007 Patel 2010 De Backer 2010	Systematic Review 6 RCTs(5 in septic shock and one in shock at large (De Backer 2010)	In-hospital or 28-day mortality	Norepinephrine is associated with lower mortality RR: 0.91; 95% CI [0.83 - 0.99]; P=0.028	The largest study (De Backer 2010) did not only include patients in septic shock (n=1679 at 28 days) (see previous comment about Cochrane review).
			Rate of arrhythmic events at 28 days	Norepinephrine is associated with lower arrhythmic events RR: 0.43; 95% CI [0.26-0.69]; P< 0.001
De Backer D, et al. 2012 Belgium	norepinephrine vs dopamine. Septic shock only. Observational trial of 1360 patients with septic shock only Martin 2000 Hall 2004 Sakr 2006 Povoa 2009 Boulain 2009. RCTs of 1408 patients in septic shock only Ruokonen 1993 Martin 1993 Marik 1994 Mathur 2007 Patel 2010 De Backer 2010(sepsis subgroup only).	2 separate Meta-analyses were conducted: one included only observational studies and the other included only RCTs. 5 observational trials in septic shock. 6 RCTs in septic shock.	28-day mortality (observational)	When all observational studies were included dopamine was not associated with a difference in mortality at 28 days RR, 1.09; 95% CI [0.84, 1.4], p =.72. After removing the Povoa et al study (due to heterogeneity), dopamine increases mortality at 28 days RR: 1.23; 95% CI [1.05–1.43]; p<.01.	In the observational studies, there is presence of heterogeneity due to the Povoa study. p <.001; I2 = 79.3%; 95%CI[50.9–91.3]. In the RCTs, the largest study included (De Backer 2010) did not only include septic shock patients. Contrary to the Cochrane meta-analysis, this meta-analysis only included the subgroup of septic shock patients (n= 1044). This increases the external validity of the results for our question about septic shock patients. However, randomization in the original De Backer RCT (2010) was not stratified according to the cause of shock. Thus, including only the subgroup of septic shock patients in this meta-analysis has the potential to induce bias.
			28-day mortality (RCT)	In RCTs dopamine increase mortality at 28 days RR, 1.12; 95% CI 1.01–1.20; p = .035

Comment(s)

Septic shock is a condition of severe infection with end-organ dysfunction and refractory hypotension. Rapid fluid resuscitation and early antibiotic therapy is essential. Vasopressor and inotropic support is also often required to maintain end-organ perfusion. In Emergency Departments, norepinephrine and dopamine are often used to treat patients in septic shock. In the last few years, many new studies have compared these two vasoactive drugs. Six majors RCTs were analysed by three different groups of systematic review authors. The Cochrane review group revised vasopressors in all types of shock. They found an increase in arrhythmias associated with the use of dopamine. No increase in mortality was found at 12 months but a trend in disfavour of dopamine can be noted. In the other two meta-analyses (De Baker and Vasu); an increase in mortality was found at 28 days. The largest data for these 2 meta-analyses came from a septic shock subgroup of patients (n=1044) from a RCT (De Baker 2010). Although the outcome analysis of this subgroup was planned by the authors in the original RCT, randomisation was not stratified for septic shock, thus inducing a certain risk of bias because of unequal confounding factors. De Baker also conducted a systematic review with 5 observational studies. No increase in mortality was found with dopamine compared with norepinephrine. When excluding a study that was the source of heterogeneity, an increase in mortality was noted at 28 days. A decrease in arrhythmias with norepinephrine was also noted in the Vasu study. Arrhythmias have the potential to impair cardiac function, leading to poor outcome.

Clinical Bottom Line

In shock, use of dopamine causes more arrhythmias than norepinephrine. In 2 out of 3 systematic reviews, dopamine increases mortality at 28 days when compared to norepinephrine. When outcomes are measured at 12 months in patients with all causes of shock, there remains a trend toward an increase in mortality with the utilisation of dopamine compared to norepinephrine. Norepinephrine seems superior compared to dopamine and should be used in septic shock.

References

1. Havel C, et al. Vasopressors for hypotensive shock (Review). Cochrane. Cochrane Database of Systematic Reviews 2011, Issue 5
2. Tajender S. Vasu, et al. Norepinephrine or Dopamine for Septic Shock : Systematic Review of Randomized Clinical Trials J Intensive Care Med 2012, 27(3): 172-78
3. De Backer D et al. Dopamine versus norepinephrine in the treatment of septic shock: A meta-analysis* Crit Care Med 2012 Vol. 40, No. 3, 725-30