Summary of findings 3. Low cardiac output syndrome.

Estimate of effects, confidence intervals, and certainty of the evidence for prevention of LCOS for paediatric patients undergoing surgery for congenital heart disease
Population: paediatric patients undergoing surgery for congenital heart disease Interventions: milrinone, milrinone low dose, levosimendan Comparator (reference): placebo or combination of milrinone + dopamine Outcome: prevention of LCOS Setting: inpatient								Figure 3
Total studies: 5 Total participants: 513 Total events: 85	Relative effect (95% CI)	Anticipated absolute effect (95% CI)			No. of participants (studies)	Certainty of the evidence	Ranking (SUCRA*)	Comments
		Without intervention	With intervention	Difference
Levosimendan	RR 0.45 (0.24 to 0.83) Network estimate	267 per 1000a	120 per 1000	147 fewer per 1000 (203 fewer to 45 fewer)	144 (4 RCTs)	High Due to heterogeneity, but large magnitude of effect	0.80	Levosimendan results in a large reduction in LCOS, compared to placebo
Milrinone	RR 0.46 (0.24 to 0.89) Network estimate	267 per 1000a	123 per 1000	144 fewer per 1000 (203 fewer to 29 fewer)	122 (4 RCTs)	Moderate Due to within‐study bias and heterogeneity, but large magnitude of effect	0.77	Milrinone likely results in a large reduction in the incidence of LCOS, compared to placebo
Milrinone low‐dose	RR 0.70 (0.39 to 1.28) Network estimate	267 per 1000a	187 per 1000	80 fewer per 1000 (163 fewer to 75 more)	79 (1 RCT)	Low Due to within‐study bias and imprecision	0.39	Milrinone low dose may reduce the incidence of LCOS, compared to placebo
Placebo	RR 1	Not estimable	Not estimable	Not estimable	168 (2 RCTs)	Reference comparator	0.05	‐
Heterogeneity: tau2 = 0; I2 = 0% (95% CI 0% to 89.6%). Q total 1.2 (df = 2, P = 0.55) Q within designs 0.05 (df = 1, P = 0.82) Q between designs 1.14 (df = 1, P = 0.29)

*SUCRA: surface under the cumulative ranking curve, calculated with R package netmeta

df: degrees of freedom

GRADE Working Group grades of evidence (or certainty in the evidence)

High certainty: we are very confident that the true effect lies close to that of the estimate of the effect.

Moderate certainty: we are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.

Low certainty: our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect.

Very low certainty: we have very little confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect.

Explanatory footnotes

^aBaseline risk (assumed control risk) obtained from Hoffman 2003, placebo group.

Certainty of the evidence for levosimendan was downgraded one step due to heterogeneity, but upgraded one step due to large magnitude of effect, resulting in 'High'.

Certainty of the evidence for milrinone was downgraded two steps due to within‐study bias and heterogeneity, but upgraded one step due to large magnitude of effect, resulting in 'Moderate'.

Certainty of the evidence for milrinone low dose was downgraded two steps due to within‐study bias and imprecision, resulting in 'Low'.

Abbreviation(s)
CI: confidence interval; df: degrees of freedom; LCOS: low cardiac output syndrome; No.: number; RCT: randomised controlled trial; RR: risk ratio; SUCRA: surface under the cumulative ranking curve.