|
Outcomes (Number of Studies)
|
Events in kidney transplant patients
|
Events in non-transplanted patients
|
Certainty of evidence (GRADE)
|
Rationale
|
|---|
|
Risk of bias
|
Inconsistency
|
Indirectness
|
Imprecision
|
Publication bias
|
Strong association
|
|---|
|
30 day/In-hospital mortality (8)
|
449/8151
|
79,978/3,272,453
|
Low a,d,g,f
|
↓
|
–
|
–
|
–
|
–
|
↑
|
|
Myocardial infarction (4)
|
104/4793
|
55,029/3,088,295
|
Very low a,b,c,g
|
↓
|
↓↓
|
–
|
↓↓
|
–
|
–
|
|
Stroke (7)
|
91/7636
|
54,508/4,794,679
|
Very low a,b,c,g
|
↓
|
↓↓
|
–
|
↓↓
|
–
|
–
|
|
Cardiac failure (3)
|
1109/7129
|
126,004/1,848,386
|
Very low a,c,g
|
↓
|
↓↓
|
–
|
–
|
–
|
–
|
|
Surgical site infection (8)
|
293/8360
|
85,747/3,536,106
|
Very low a,b,c,g
|
↓
|
↓↓
|
–
|
↓
|
–
|
–
|
|
Pneumonia (8)
|
693/8151
|
196,982/3,272,453
|
Very low a,c,g
|
↓
|
↓↓
|
–
|
–
|
–
|
–
|
|
Sepsis (5)
|
692/4019
|
15,951/164,139
|
Low a,c,g,f
|
↓
|
–
|
–
|
–
|
–
|
↑
|
|
Gastrointestinal events (6)
|
495/8006
|
414,890/3,272,004
|
Very lowa,c,g,h
|
↓
|
↓↓
|
–
|
↓
|
–
|
–
|
|
Blood transfusion (5)
|
2056/7505
|
424,174/3,087,470
|
Very lowa,c,g,h
|
↓
|
↓
|
–
|
–
|
–
|
–
|
|
Unplanned return to theatre (6)
|
53/395
|
75/1524
|
Very lowa,b,g
|
↓
|
↓
|
–
|
↓↓
|
–
|
–
|
|
AKI (11)
|
2205/15,371
|
316,667/7,807,595
|
Moderate a,c,e,g
|
↓
|
↓↓
|
–
|
–
|
–
|
↑↑
|
|
PE/DVT (3)
|
204/4723
|
81,226/3,087,400
|
Very low a,b,c,g
|
↓
|
↓↓
|
–
|
↓↓
|
–
|
–
|
- aCertainty of evidence downgraded due to concerns of poor comparability between study cohorts
- bCertainty of evidence downgraded for imprecision, due to few events and wide confidence intervals
- cCertainty of evidence downgraded for inconsistency due to significant heterogeneity that remained unexplained (ranged between 67% and 98%)
- dCertainty of evidence not downgraded due to significant heterogeneity being explained by surgical disciplines and higher prevalence of diabetes in kidney transplant patients
- eCertainty of evidence upgraded due to large (>5 fold) risk estimate
- fCertainty of evidence upgraded due to large (>2 fold) risk estimate
- g Publication bias not judged to assess certainty of evidence due to small study numbers (<10)
- hCertainty of evidence not downgraded for imprecision due to significant number of events in both groups
