SYSTEMATIC REVIEW

Criteria abstracted from The Users' Guides to the Medical Literature series in JAMA

Ann Intern Med. 2005 Apr 5;142(7):510-24.[abstract]

Review posted December 19, 2006

1. Are the results of the study valid?
   
   
   
   
   1. Primary questions:
      
      
      
      1. Did the overview address a focused clinical question?
         
         Not really. The primary aim of the overview was to evaluate and provide a broad – not focused - perspective of the effects of low dose dopamine (≤ 5mcg/kg/min), compared with placebo or no therapy in patients with or at risk for acute renal failure. The outcome measures were clearly stated by the authors. The primary outcome was mortality and secondary outcomes were a) requirement for renal replacement therapy, b) renal physiologic variables (urine output, serum creatinine level, or measured creatinine clearance on days 1, 2, or 3 after starting therapy), or c) adverse effects.
      
      
      2. Were the criteria used to select articles for inclusion appropriate?
         
         Yes the criteria for inclusion appear appropriate. Two reviewers conducted independent search strategies and selected randomized and quasi- randomized controlled trials of low dose dopamine versus control. They had strict criteria for methodologic assessment of allocation and concealment of randomization schedule, blinding of caregivers and outcome assessors, and the number and reasons for post randomization withdrawals.
      
      
      
      
   
   
   2. Secondary questions:
      
      
      
      1. Is it unlikely that important, relevant studies were missed?
         
         Yes. The authors did a comprehensive search of five databases MEDLINE (1966- Jan 2005), EMBASE (1980-2005), CANCERLIT (1975-Oct 2002), CINAHL (1982-Jan 2005), and CENTRAL without age or language restrictions. They also searched the Renal Health Library on Feb 3, 2005. Abstract proceedings, reference lists and personal files were searched for potential randomized control trial (RCT) citations. The authors also attempted to minimize publication bias by contacting experts in the field to identify any unpublished studies. Furthermore funnel plots of the relevant data also did not suggest the presence of publication bias. However they did not systematically search “grey literature� (such as conference proceedings). This leaves a chance of still missing unpublished trials and articles. Overall the systematic review identified 61 randomized and quasi randomized controlled trials enrolling 3359 patients.
      
      
      2. Was the validity of the included studies appraised?
         
         Methodological quality of the included studies was assessed according to use of intention-to-treat analysis and reported allocation generation and concealment of randomization schedule, as well as reporting of complications and co-interventions. Ideally studies could be scored to determine their validity. However quality scores were not derived making it difficult to appraise studies in an objective manner. Randomization was not concealed in one study and not reported in 12 studies. The treatment was not blinded in 22 studies, and was not described in four studies. Only 5 studies included 100 or more patients. In some of the studies there was limited data available on the use of therapies other than dopamine and not all studies incorporated specific treatment protocols. Typically reviewers will conduct sensitivity analyses , eliminating lower quality trials to see if the effect size has changed; that was not done in this study.
      
      
      3. Were assessment of studies reproducible?
         
         Yes. The authors addressed the issues of reproducibility of assessment, and two independent authors were involved in the study selection, methodologic quality assessment, and data abstraction. Disagreements between the reviewers were resolved by consensus. The two reviewers achieved excellent agreement on selection of trials for inclusion (k=0.98 [95% CI, 0.96 to 1.00] ).
      
      
      4. Were the results similar from study to study?
         
         Results were not similar from study to study. Appropriate statistical methods were used to assess for heterogeneity. Meta analysis was done using random-effects model to aggregate data for each outcome, and to assess treatment effects. Heterogeneity was further characterized as significant or insignificant based on extent (Q statistics), with a p value of 0.10 or less indicating significant heterogeneity and also based on impact ( I²), where substantial heterogeneity exists when I² exceeds 50%. After data analysis it was found that effect of low dose dopamine on renal physiologic outcome showed substantial heterogeneity with I² values ranging from 36-94%. Disparate patient population, many subgroups with few trials, illness severity, cointervention and study method may explain heterogeneity. It is also possible that trials without allocation concealment or blinding inflate treatment effects. However there was no evidence of between study heterogeneity for outcomes such as mortality and renal replacement therapy.
      
      
      
      
   
   
   
   


2. What are the results?
   
   
   
   
   1. What are the overall results of the review?
      
      61 randomized and quasi- randomized trials assigned 3359 patients were identified.
      
      Meta analysis using random- effect models showed the following results.
      
      Effect of low dose dopamine on clinical outcomes:
      
      The overall results are that low dose dopamine does not seem to alter the outcome of interests in this systematic review. Neither mortality nor the need for renal replacement therapy was impacted by the use of this intervention.
      
      54 trials which reported mortality and renal replacement therapy; the pooled analyses showed no effect of low dose dopamine on mortality (relative risk, 0.96 [95%CI, 0.78 -1.19]) or the need for renal replacement therapy (relative risk, 0.93; 95%CI, 0.76-1.15). There was no evidence of between study heterogeneity for these outcomes.
      
      Adverse effects:
      
      Overall 50 trials reported adverse effects. Patients in the ANZICS trial (90) accounted for most of adverse effects which included myocardial infarction ( 4 pts receiving dopamine and 2 control pts), cutaneous ischemia from extravasation injury (2 pts receiving dopamine) and worsening of Raynaud’s disease (1 pt receiving dopamine). The remaining pts had tachyarrythmias. Pooled analysis data did not demonstrate a statistically significant difference in adverse events between low dose dopamine and control groups (relative risk, 1.13; 95% CI, 0.90- 1.41). However possibility regarding underestimation in final analysis may not be excluded as the authors have duly pointed out.
      
      Effect of low dose dopamine on renal physiologic outcomes:
      
      Pooled data analysis showed an increase in urine output by 24% (ratio of means of urine output 1.24; 95% CI, 1.14 -1.35; P< 0.001) on day 1. Improvement in serum creatinine level (4% relative decrease; 95%CI, 1% to 7% ) and measured creatinine clearance (6% relative increase; 95%CI, 1% -11%) on day 1 were clinically insignificant. There were no significant changes on day 2 and 3 of therapy.
      
      The ANZICS trial (90) was not included in the pooled urine output analyses because it measured hourly urine out puts at specified times rather than cumulative urine output. However additional data provided by the ANZICS investigators also showed a trend towards increased hourly urine output at 24 hrs (corrected for baseline urine output) in the dopamine group compared to placebo group. However at 48 hrs hourly urine output did not differ. There was however very significant heterogeneity with I² values ranging from 36% to 94% for 8 of the nine physiologic outcomes.
   
   
   2. How precise were the results?
      
      Fairly precise. The confidence intervals of each of the results are mentioned above. The confidence limits are not wide in terms of mortality and renal replacement therapy. Most CI's cross or nearly cross zero (or one for RR), suggesting that the differences between study groups are not statistically significant. The lack of effect on important clinical outcomes was consistent across trials
   
   
   
   


3. Will the results help me in caring for my patients?
   
   
   
   
   1. Can the results be applied to my patient care?
      
      This meta-analysis showed no effect of the intervention on either survival or renal function. The application of this evidence to pediatric population needs to be done with caution. The incidence and pathophysiology of death and renal dysfunction are different between critically ill adults and children. Without a well designed RCT in pediatric population to prove its benefit, low-dose dopamine should not be either recommended or completely excluded for the use in critically ill patients with risk of death or renal dysfunction.
   
   
   2. Were all clinically important outcomes considered?
      
      Yes, the authors accounted for clinically relevant outcomes including all cause mortality, adverse effects, requirement for renal replacement therapy, renal physiologic variables ( urine output, serum creatinine level, or measured creatinine clearance on days 1, 2, and 3 after starting therapy.
   
   
   3. Are the benefits worth the harms and costs?
      
      No. This study did not show any benefit from the use of low-dose dopamine on clinically relevant outcomes. There was lack of good evidence, that low-dose dopamine offers important clinical benefits to patients with or at risk for acute renal failure. The transient improvement in renal physiology manifested by increase in urine out put on day 1 was not clinically significant. The treatment effect was also small. Therefore it is probably not worth any harm and costs.
   
   
   
   

Last Updated: December 19, 2006

Fill out our Feedback Form | Editorial Notes | About Us