I have three recommendations for hemostasis during surgery:
In general, I recommend conventional treatment (Shape, Clean, Pack) first, followed by a 30-day check to see if the fistula has closed. If the sinus tract persists, then a regimen of antibiotics. If the sinus tract persists further, then a surgical approach may be considered.
I routinely use 2-4 carpules of 1:50,000 xylocaine when performing surgery. Block anesthesia gives you patient comfort, but is not so good at hemostasis; therefore, infiltrating in the apical regions of the teeth included in the incision will provide powerful hemostasis and enhanced vision during surgery. In 25 years I have found this to be most effective. Obviously, certain patients' metabolic rate necessitates supplemental support anesthesia which tends to never be as effective as the original battery of injections. I use Marcaine (1:200,000) infrequently for out-of-town patients following suturing to bridge the time interval between the surgery and the patient's return to their home. Since the adrenal medulla produces endogenous epinephrine, there's no such thing as a patient allergic to adrenaline, although there may be some undesirable non-threatening cardiac events.
Using abrasive coated tips (Dentsply Tulsa Dental) goes a long way towards eliminating gutta percha adhering to prepped walls. Additionally, prep slightly deeper than your intentions so when and if you need to condense gutta percha coronally, the depth of your prep is ideal. For packing EBA, I make all of my own pluggers from old endo explorers and customize them to address any clinical situation.