More info from Dick Shevell's AIAA paper "Aerodynamic Bugs: Can CFD Spray Them Away"
"DC-10 wind tunnel tests showed a significant loss in maximum lift
coefficient in the flap deflected configurations, with landing slat
extension, compared to predictions. This resulted in a stall speed
increase of about 5 knots in the approach configuration. The initial
wing stall occured behind the nacelles and forward of the inboard
ailerons. The problem was traced by flow visualization techniques to the
effects of the nacelle wake at high angles of attack and the absence of
the slat in the vicinity of the nacelle pylons. The solution was
developed in the NASA Ames Research Center 12 ft. pressurized tunnel and
turned out to be a pair of strakes mounted forward on each side of the
nacelles in planes about 45 degrees above the horizontal. The final
strake shape was optimized in flight tests. The strakes are simply
large vortex generators. The vortices mix the nacelle boundary layer air
with the free stream and reduce the momentum loss in the wake. The
vortices then pass just over the upper surface of the wing, continuing
this mixing process. The counterrotating vortices also create a downwash
over the wing region unprotected by the slat, further reducing the
premature stall. The effect of the strakes is to reduce the required
takeoff and landing field lengths by about 6%, a very large effect."
http://www.aiaa.org/content.cfm?pageid=406&gTable=mtgpaper&gID=43846