Quality assessment:
67% (N=8) of the included studies were classified as low risk for external validity. Only 18% (N=2) of the studies were low risk for internal validity, mostly due to the predominance of retrospective studies. All studies included consecutive recruitment over a time frame.