1. Patient Selection and Disclosure Emily Finlayson, MD, MS Department of Surgery University of California, San Francisco

2. What we’re going to cover Mortality after surgery in the elderly –Fact v Fantasy Recovery after surgery –Longer than your surgeon said it was going to be What patients value –Not always the same as your family or your surgeon 2

3. Context Population is aging –274 million  352 million –13% of population  20% of population An increasing number of very elderly patients will be candidates for major surgery Are these patients undergoing surgery? 3

4. It’s a cancer, so it has to come out, right? 4

5. Some Decisions are Pretty Easy 5

6. Some Decisions Are Pretty Easy 6

7. Other Decisions Are Not So Easy 7

8. Are Older Patients with Cancer Undergoing Surgery? 8

9. O’Connell et al, Ann Surg Oncol, 2004

10. Assumptions Surgery in the elderly is getting safer ‘Esophageal resection for carcinoma in patients older than 70 years old.’ Ann Surg Oncol. 2002;9(2):210-214. ‘Pancreaticoduodenectomy in the very elderly.’ Jour GI Surg. 2006;10(3):347-56.

11. Are These Results Generalizable? Selective submission, publication bias Consider the source –Centers of Excellence Trial data –Sick and elderly patients often excluded “Real world” mortality and survival data –The ‘benefits’ side of the equation 11

12. National Benchmark Data: Mortality after Major Cancer Surgery Retrospective cohort study of patients 65+ undergoing major cancer resections (n=14,088) –Lung –Esophageal –Pancreas SEER-Medicare (1992-2001) Outcomes –Operative mortality –5-year survival Finlayson et al, J Am Coll Surg, 2007

13. 13

14. 14 If Elderly Cancer Patients Make It Through Surgery, Do They Survive Long Term?

15. 15

16. 16 Comorbidity Counts Cancer5 year survival (%) Lung <2 comorbidities37 2+ comorbidities28 Esophagus <2 comorbidities21 2+ comorbidities17 Pancreas <2 comorbidities18 2+ comorbidities5

17. 17 National Benchmark Data: Discharge Disposition Retrospective cohort study of patients undergoing major cancer resections (N= 601,081) –Lung –Esophageal –Pancreas Nationwide Inpatient Sample (1994-2003) –Discharge disposition stratified by age Finlayson et al, J Am Coll Surg, 2007

18. Discharge to SNF after Surgery, by age 18 OperationAge 65- 69 Age 70- 80 Age 80+ Lung resection4%8%16% Pancreatectomy8%16%24% Esophagectomy6%12%30%

19. OK…but those are big operations. What about the bread and butter stuff? 19

20. GI surgery in NH Residents NH residents 65+ undergoing GI surgery in the US Medicare inpatient file + MDS (1999-2006), N=70,719 –Bleeding DU –Benign colon disease –Cholecystitis –Appendicitis Operative mortality compared to 1.1 million Medicare beneficiaries 65+ 20 Finlayson et al, Ann Surg, 2011

21. Outcomes of Interest Operative mortality Secondary interventions –Mechanical ventilation > 96 hrs –Central venous catheterization –PA catheter placement –IVC filter placement –Bronchoscopy –Feeding tube placement –Tracheostomy placement 21 Finlayson et al, Ann Surg, 2011

22. 22 Operative Mortality Finlayson et al, Ann Surg, 2011

23. 23 Any invasive intervention (%) Diagnosis NH Resident General Population Bleeding DU Survivors42.236.2 Deaths63.061.2 Benign colon Survivors40.722.4 Deaths56.854.6 Cholecystitis Survivors15.04.5 Deaths40.736.0 Appendicitis Survivors18.35.5 Deaths40.3 43.2 Finlayson et al, Ann Surg, 2011

24. What other choice do we have? Life and death situations…. Consider alternative therapies in patients with limited life expectancy –Antibiotics –Cholecystostomy tube –Colonic stents –IR for bleeding 24

25. What do we know about the trajectory of recovery after major surgery?

26. Functional Status after Surgery 372 patients age 60+ Elective major abdominal operations (GS, GYN) Functional assessments –Preoperative –1, 3, and 6 weeks, 3 and 6 months Lawrence et al, J Am Coll Surg, 2004

27. 27

28. 28

29. 29

30. 30

31. What about functional recovery in the very frail?

32. Functional Outcomes in NH Residents NH residents 65+ undergoing colectomy for cancer Medicare inpatient file + MDS (1999-2006), N=6822 Functional trajectories after surgery –MDS-ADL score (0-28) 1 year mortality 32 Finlayson et al, JAGS, in press

33. Functional trajectories and 1 year morality 33 Finlayson et al, JAGS, in press

34. Functional trajectories and 1 year morality, stratified by baseline function 34 Finlayson et al, JAGS, in press

35. ADL decline, maintenance of ADL, and death 35

36. 36 Characteristic% declinedRR, 95% CI Age 80+52.8 1.53 (1.15-2.04) Pre-op decline59.9 1.21 (1.11-1.32) Hospital readmission51.8 1.15 (1.03-1.29) Surgical complication55.3 1.11 (1.02-1.21) Urgent admission52.5 1.10 (1.03-1.18) Finlayson et al, Ann Surg, 2011 Predictors of Functional Decline

37. What outcomes are really valued by older patients with limited life expectancy?

38. Treatment Preferences in Patients with Limited Life Expectancy 226 subjects with limited LE given hypothetical scenarios Burden of treatment –LOS, testing, invasive procedures Expected outcome –Restoration of current health –Death –Functional impairment –Cognitive impairment Fried et al, N Engl J Med, 2002

39. 39 Treatment Intensity Health Outcome Wants treatment Low BurdenReturn to Current Health 98.7% High BurdenReturn to Current Health 88.9% Low BurdenFunctional Impairment 25.6% Low BurdenCognitive Impairment 11.2%

41. There are Important Differences Between Decisions Made by Elder Patients and Their Surrogates 41

42. Patient-Surrogate Agreement about Acceptable Outcomes >80% for health states –Current health, mild memory impairment –Coma 61-65% for severe pain –Patients/surrogates equally likely to rate as acceptable 58-62% for severe functional impairment –Surrogates more likely to rate as acceptable 42 Fried et al, Arch Intern Med, 2003

43. How Can We Improve Surgical Care in Frail Elders? 43

44. Developing Quality Indicators for Elderly Surgical Patients RAND/UCLA project –Expert panel from surgery, geriatrics, anesthesia, critical care, internal, and rehabilitation medicine –Formally rated the indicators using a modification of the RAND/UCLA Appropriateness Methodology –Identified 91 candidate indicators rated as valid 44 McGory et al, Ann Surg, 2009

45. Developing Quality Indicators for Elderly Surgical Patients 6 Domains Unique to Elderly Patients –Comorbidity assessment –Evaluation of elderly issues –Medication use –Patient-to-provider discussions –Postoperative management –Discharge planning 45 McGory et al, Ann Surg, 2009

46. Elderly-Specific Process Measures Patient-to-provider discussions –Assess patient’s decision-making capacity –Specific discussions on expected functional outcomes –Advanced directives: life-sustaining preferences, surrogate decision maker –Clarify goals of care 46 McGory et al, Ann Surg, 2009

47. Summary Nationwide, operative mortality remains high and survival is low among the very elderly undergoing major cancer surgery Even for less complex procedures, mortality is very high in frail patients Functional recovery after major surgery is protracted in elders Patients with poor prognosis value function, cognition, and quality of life very highly

48. Implications Comprehensive assessment –Medical –Functional –Cognitive Realistic expectations essential for true informed consent Need for multidisciplinary approach, care pathways for geriatric patients 48