Impact of hospital volume on outcomes for pancreaticoduodenectomy: a single UK HPB centre experience.

BACKGROUND

High hospital volume has a favorable impact on outcomes for complex procedures including pancreaticoduodenectomy (PD); however, the temporal relationship has not been evaluated in a single centre.

AIM

To evaluate the impact of UK cancer outcome guidelines (COG) on outcomes for PD in a single UK HPB specialist centre.

PATIENTS AND METHODS

All patients with pancreatic pathologies undergoing surgery at our institution from 1999 to 2006 were identified, of which 140 underwent PD. The annual caseload for PD and corresponding outcomes for length of hospital stay, morbidity, mortality and survival were analysed during the period around the implementation of UK COG with an increase in the surgical workload correlating with catchment's population increase from 1.6 to 3.1 million.

RESULTS

Between January 1999 and December 2006, 140 patients underwent a PD (M:F 1.06:1; median age 64 (range 34-84) years). Median hospital stay was 16 days (range 7-318). The 30-day mortality was 2.8%, in-hospital mortality was 6.4% and morbidity was 37.1%. Pancreatic leak/fistula rate was 8.6%. Over the 7-year period, PDs per year increased 5.3 fold from 6 procedures in 1999 to 32 in 2006. Analysis of the data for 1999-2002-(pre-COG) and 2003-2006-(post-COG) showed a trend towards decrease in mortality (from 9.7% to 5.0%, p = 0.448: OR = 2.74 (95% CI, 0.58-12.88); Fisher's exact test) and morbidity (from 41.6% to 35.3%; OR = 1.29 (95% CI, 0.74-3.56); p = 0.565).

CONCLUSION

With COG implementation within a single UK pancreatic unit, the PD volume and staffing levels increased with a trend towards decreased morbidity and mortality.