*Manuscript (Please include all author details) Click here to download Manuscript (Please include all author details): Article clean copy.doc Click here to view linked References 1 2 Late survival in non-operated patients with infra-renal abdominal aortic aneurysm 3 4 5 Scott SWM, Batchelder AJ, Kirkbride D, Naylor AR, Thompson JP Departments of Vascular Anaesthesia and Vascular Surgery, Leicester Royal Infirmary 6 1
7 What does this study/review add to the existing literature and 8 how will it influence future clinical practice 9 10 11 12 13 14 The risk of rupture in patients with 5.5-6.9cm AAAs may not be as high as previously thought, suggesting that there is ample time to optimise co-morbidities prior to deferred surgery or to advise that it may be safer to offer elective surgery at a higher diameter threshold in higher risk patients. Conversely, the risk of rupture in patients with AAAs >7cm is very high, thereby mandating a much more aggressive approach to intervening, even in higher risk patients. 15 16 2
17 Abstract 18 Objectives 19 20 Historical studies report high rupture rates in patients with non-operated >5.5cm abdominal aortic aneurysms (AAA), although a recent audit has questioned this. 21 Methods 22 23 24 25 26 Retrospective review of 138/764 AAA patients (18%) evaluated in a pre-assessment anaesthetic clinic (PAC) between 2006-2012, who either did not undergo elective AAA repair or who underwent deferred repair. The remaining 626 underwent repair. Patients with severe co-morbidities (dementia, advanced malignancy, life-expectancy <1 year) and not referred to PAC were excluded. 27 Results 28 29 30 31 32 33 34 At a median of 27 months; 71 (52%) died; 36 (51%) following rupture. Cumulative survival, free from rupture or surgery for acute symptoms was 96% at 1-year, 84% at 3-years and 64% at 5-years, where baseline AAA diameters were 5.5-6.9cm. For diameters >7cm; survival, free from rupture was 65% at 1-year, 29% at 3-years and 0% at 5-years. Median interval to rupture was 47 months (5.5-6.9cm AAAs) and 21 months where baseline diameters were >7cm. Rupture accounted for 32% of late deaths in 5.5-5.9cm AAAs, 46% (6.0-6.9cm) and 71% for AAAs >7cm. 35 Conclusions 36 37 Approximately half of all late deaths in this non-operated cohort were not AAA related, suggesting that even had repair been undertaken, it would not have prolonged patient 3
38 39 40 41 42 43 44 survival. The incidence of rupture in high-risk patients with an AAA <7cm diameter was less than 5% at 12 months, thereby giving ample time to optimise risk factors and improve preexisting medical conditions prior to undertaking a deferred intervention. Even if these patients do not undergo surgical repair, the risk of late rupture was relatively low. By contrast, non-operated patients with >7cm AAAs face a very high risk of rupture and will probably benefit from elective surgery, with the caveat that a higher procedural risk might have to be incurred. 45 46 4
47 Introduction 48 49 50 51 52 53 54 55 56 57 Aortic abdominal aneurysm (AAA) is an important cause of mortality, predominantly in men. The mortality rate following AAA rupture is approximately 80% and 8000 people in the UK and 15000 people in the USA die each year from aneurysm rupture 1. The risk of rupture increases with the size of aneurysm and its rate of growth. Data from 2002 suggested a 9% risk of rupture for 5.5-5.9cm aneurysms within one year, compared with 19% for 6.0-6.9cm aneurysms and 33% for aneurysms greater than 7cm in diameter 2. A UK study of 106 patients with AAA larger than 5.5cm who had been turned down for elective surgery reported an overall three-year survival rate of just 17%, with ruptured AAA being responsible for death in 36%, 50% and 55% of patients with AAA diameters of 5.5-5.9cm, 6-7cm and >7cm respectively 3. 58 59 60 61 62 63 64 65 66 67 68 Median survival times for untreated AAAs have also been reported to be poor and are often cited to justify surgical intervention. Historical data from 1998 suggested a median survival of only 18 months in patients declined surgical repair (21 months in patients whose AAA diameters were 5-5.9cm and 15 months where the AAA diameter was >6cm) 4. However, more recently published data suggest that survival rates may be higher. A 2013 audit of survival rates in 72 AAA patients who had been turned down for elective repair reported median survival times of 44 months in patients with an aneurysm diameter of 5.1-6.cm, compared with 26 months for AAAs up to 7cm in diameter, but only 6 months median survival for AAAs >7cm 5. Importantly, however, endovascular aneurysm repair (EVAR) was not offered to any of these patients. 5
69 70 71 72 73 74 75 76 77 78 79 80 81 In order to ensure optimal work-up of patients being considered for elective AAA repair, a vascular anaesthetic pre-assessment clinic (PAC) was established in 2005 at Leicester Royal Infirmary, in accordance with guidance from the Royal College of Anaesthetists (RCoA) 9 and the Vascular Society of Great Britain and Ireland (VSGBI) 10. The majority of patients under consideration for elective AAA surgery were referred for anaesthetic pre-assessment, unless there were no significant co-morbidities that might otherwise preclude expedited surgery. Following assessment by a consultant vascular anaesthetist and subsequent surgical review, patients either: (i) consented to undergo open repair/evar; (ii) were considered fit for surgery, but the patient then declined any intervention; or (iii) the patient did not undergo immediate elective surgery, either because they were deemed unfit for open repair and EVAR was not anatomically feasible, or because further investigations or referral to another specialty was deemed necessary. 82 83 84 85 86 The aim of the current audit was to establish cumulative survival rates in patients not undergoing immediate elective AAA repair following PAC assessment, in order to establish: (i) overall survival; (ii) causes of death during follow-up; and (iii) cumulative freedom from AAA rupture or emergency AAA intervention. 87 88 6
89 Methods 90 91 92 93 94 95 96 97 98 All patients with an AAA >5.5cm and who were under consideration for elective AAA repair were first evaluated in the vascular surgery clinic. A relative minority of patients with very major medical co-morbidities that precluded any type of operative intervention (e.g. severe dementia, advanced malignancy or any medical condition with a life expectancy <1 year) were advised against surgical intervention following a full discussion of risks and benefits with patients and/or their relatives. Unfortunately, no record was maintained of the number of these patients. The remainder (majority) were then referred for either elective AAA repair without referral to the PAC (because they had few or no medical comorbidities) or were referred for formal PAC assessment. 99 100 101 102 103 104 105 106 107 108 109 110 A retrospective review of PAC case records between 1st January 2006 and 30th April 2012 was undertaken and the management strategy for each patient assigned to one of three subgroups: (i) patient deemed fit and then underwent either open or EVAR repair; (ii) patient deemed fit for open repair and/or EVAR, but the patient then declined intervention; and (iii) the patient did not undergo immediate elective surgery, either because they were deemed unfit for elective surgery (but not necessarily emergency repair), or because further investigations or referral to another specialty were deemed necessary before possibly undergoing elective surgery, or where a decision had been made to wait for the AAA diameter to increase to 6cm. PAC records were cross-referenced with data entered into the vascular surgery audit database in order to capture those patients who underwent elective AAA repair without referral to the PAC. 7
111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 The causes of death in patients who did not undergo immediate elective surgery after PAC assessment (i.e. groups (ii) and (iii) (see above)) were analysed at April 2013 using hospital case notes, locally held death certificates, or by the purchase of death certificates from the United Kingdom General Records Office (for patients referred from outwith Leicestershire). Data were analysed using PASW statistics version 20 (SPSS Inc, Chicago, Illinois, USA) and Excel 2010 (Microsoft, Redmond, California, USA) and Kaplan-Meier analyses were used to estimate cumulative mortality, time to delayed AAA repair or fatal rupture. Life-table analyses were used to establish estimated risks of rupture and/or delayed intervention at one, three and five years following any decision in the PAC not to undergo immediate elective AAA repair. Accordingly, the cohort under investigation included patients who were advised against elective repair who may or may not have undergone an emergency AAA intervention at a later date, patients who declined intervention and those undergoing further evaluation/investigation prior to any decision to undergo surgery. Comparisons of differences between groups were analysed using log-rank testing. 126 127 8
128 Results 129 130 131 132 133 134 135 136 137 138 Between 1 st January 2006 and 30 th April 2012, 764 patients with an AAA 5.5cm were under active consideration for elective infra-renal AAA repair (figure 1). Of these, 575 (75%) were referred and seen in the PAC, while 189 patients with few significant co-morbidities (25%) underwent elective open or EVAR repair of their aneurysm without PAC assessment. Following PAC assessment, 437 (76%) underwent elective aneurysm repair while 138 (24%) did not. The demographics of the operated and non-operated patients are summarised in table 1. The 138 non-operated patients were predominately male (115 [83%] versus 23 [17%] female), with a median age of 77 ([IQR 72-81]) years. Details regarding aneurysm sizes in patients in the non-operated group are shown in table 2. Median follow-up for the entire cohort was 27 (0-82[15-40]) months, representing 335 years of patient follow-up. 139 140 141 142 143 144 145 146 147 148 149 150 The main reasons for not proceeding to immediate (elective) surgical repair in these 138 patients were: (i) the patient was advised not to proceed with elective surgery at that time (n=71 (51%)); (ii) the patient was deemed fit enough to undergo surgery and advised to proceed (n=40 (29%)), but an early operation was not undertaken (see below); and (iii) 27 patients (20%) required further investigation or referral to another specialty before undergoing elective surgery. The group of 71 patients advised not to undergo immediate elective surgery includes some who would be considered for emergency surgery should the need arise (rupture/acutely symptomatic) and also some (usually elderly) patients who were advised to remain in serial aneurysm surveillance until the AAA exceeded 6cm before operating. Out of the 40 patients who were deemed fit to undergo surgery in the PAC, 14 (35%) were placed on an elective waiting list for surgery (table 3), one was kept in serial 9
151 152 153 154 surveillance until the AAA exceeded 6cm and then underwent elective repair, 10 (25%) were not fit for open repair, but were considered fit for EVAR but were not anatomically suitable; 8 (20%) declined an operation, six (15%) were initially considered fit for elective repair but then deemed unfit at the time of admission, while one became lost to follow-up. 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 Of the 14 patients who were placed on the waiting list for surgery (table 3), three died following AAA rupture, while one patient underwent urgent repair of an acutely symptomatic aneurysm. Six underwent a delayed elective repair, while three died of medical co-morbidities before undergoing surgery and one was removed from the waiting list because of worsening health. By April 2013, 71 patients (52%) had died and the causes of death are listed in table 4. The commonest cause was AAA rupture, which occurred in 37 patients (52%). Thirty-four patients (48%) ultimately died of a non-aaa related cause. Median survival for the entire group of 138 patients not undergoing immediate/early elective surgery was 41 months (95% CI 34-48 months). Figure 2 details cumulative survival for the 138 patients not undergoing immediate/early elective surgery, stratified for aneurysm diameter. Median survival for patients with an unoperated 5.5-6.9cm AAA was 47 months (95% CI 43-51), compared with 14 months (95% CI 2-26 months) in patients whose aneurysm was >7cm (log rand p<0.001). Interestingly, the cause of death varied considerably with aneurysm size. Only 32% of all deaths in non-operated patients whose AAA (at baseline) measured 5.5-5.9cm actually died of rupture, increasing to 46% where the AAA measured 6.0-6.9cm at baseline and 71% in patients whose AAA was >7cm. 172 10
173 174 175 176 177 178 179 180 181 182 183 Figure 3 shows cumulative survival free from rupture or intervention for an acutely symptomatic AAA, again stratified for AAA diameter. Overall, 11% of non-operated patients had ruptured or undergone an emergency intervention for an acutely symptomatic AAA at 1 year, increasing to 28% at three years and 49% at 5 years. For patients with an initial AAA diameter 5.5-6.9cm, 3.9% had ruptured or undergone an emergency intervention for an acutely symptomatic AAA at 1 year, increasing to 16% at three years and 36% at 5 years. For patients with an initial AAA diameter >7cm, 35% had ruptured or undergone an emergency intervention for an acutely symptomatic AAA at 1 year, increasing to 71% at three years and 100% at 5 years. The median interval from being seen in PAC to either rupturing or undergoing surgery for an acutely symptomatic AAA was 21 months (95% CI 6-36) in patients whose baseline AAA diameter was >7cm large AAA. 184 185 11
186 Discussion 187 188 189 190 191 192 193 194 195 196 197 198 Elective open repair and EVAR are effective strategies for preventing AAA rupture in the majority of patients. However, not all AAA patients will be fit enough to undergo surgical repair. Some may be unfit for both open repair and EVAR, others may be unfit for open repair (but be fit for EVAR), only to be found to be anatomically unsuitable for an endovascular strategy. In other patients (especially the elderly), it is not unusual to recommend delaying elective surgery until the AAA reaches 6cm in diameter. As a consequence, a proportion of AAA patients will be advised to either not undergo elective repair (but still be considered for emergency surgery at a later date), or to never undergo surgery (even emergency). Unfortunately, there is a strong public perception that AAAs are synonymous with time bombs and there is a parallel perception amongst professionals that cumulative rates of rupture are very high, thereby justifying a potentially higher risk intervention. 199 200 201 202 203 204 205 206 207 208 Accordingly, the decision to recommend elective AAA repair can frequently be difficult. Moreover, the move towards individual surgeon based outcome reporting after elective AAA repair (now done yearly in the UK) may promote risk averse behaviour amongst vascular surgeons, which could mean that even more patients might be turned-down for elective surgery in the future 12. For some critics, risk averse behaviour might entail exposing too many patients to an unacceptably high risk of rupture if declined elective surgery 13, but the current study suggests that the risk of rupture may not be as high as previously thought (especially in patients with 5.5-6.9cm AAA). At the very least, there does appear to be more time than previously thought to either optimise patient risk factors or to 12
209 210 refer the patient to a more experienced vascular centre. Having more time to optimise risk factors was also one of the hidden messages in the much-criticised EVAR 2 trial 6. 211 212 213 214 215 216 217 218 219 220 221 The current study observed that patients who did not undergo early elective AAA repair had longer survival rates than previously thought; with a median survival of over 3 years for patients whose baseline AAA diameter was <7cm. For these patients, median survival free from rupture or acute surgical intervention was over 5 years. It is accepted, however, that patients whose AAA was >7cm (at baseline) face a very high risk of rupture in the first 12 months (35% in the current study). Because of this, a more aggressive approach to repairing AAAs that are >7cm is warranted, which may include referral to another centre with a greater experience of more complex endovascular repair strategies (eg fenestrated/branched EVAR, chimney/snorkel EVAR. It would also seem reasonable to request a second opinion before declining any patient with a >7cm AAA from elective repair. 222 223 224 225 226 227 228 The data from the current study corroborate the findings from a recent British study involving 72 patients deemed unfit for open AAA surgery, where median survival in these non-operated patients was 44 months, 26 months and 6 months for patients whose baseline AAA diameters were 5.1-6.0cm, 6.1-7.0cm >7.0cm AAA respectively 5. It is likely that patients survive longer than previously observed, possibly because of advances in comorbidity management. 229 13
230 231 232 233 234 235 236 237 238 This study has also shown that co-morbidities were an important cause of late mortality in patients, particularly in those patients with smaller AAAs. Aneurysm rupture was the cause of death in approximately half of all the patients in this series. As expected, this proportion varied with AAA size, but it was only when AAAs were >7cm that rupture became the predominant cause of death. Furthermore, not all of the 138 patients in this series were excluded from early elective repair on the grounds of ill-health (table 3). Eight patients were found to be fit and advised to undergo elective surgery but subsequently declined. It is therefore possible that the current cohort of patients were fitter than those recruited into EVAR II, where co-morbidities were responsible for 60% of deaths 6. 239 240 241 242 243 244 245 246 247 248 249 250 251 252 This study does, however, have limitations. This was a single-centre, retrospective study, although it remains one of the largest to look at cumulative rupture rates in non-operated patients with large AAAs. For a great many patients, the cause of death was obtained from death certificates, which can be subject to error in the absence of post-mortem examinations. Accordingly, it might be suggested that some patients with sudden death might have been falsely attributed to an acute coronary event instead of ruptured AAA. However, it might also be reasonably argued that patients known to have an AAA might be more likely to have ruptured AAA put on the death certificate in the event of a sudden death. Secondly, this study only analysed outcomes in non-operated patients who had undergone PAC assessment. We were unable to identify how many patients with severe dementia, advanced malignancy or a life expectancy <1 year were declined AAA surgery without referral for PAC assessment. It is likely, however, that had this additional cohort of severely co-morbid patients been included in cumulative survival statistics, a greater 14
253 254 255 256 257 proportion of patients would have died from non-aaa related causes. It is also possible that some patients who were declined elective surgery at Leicester Royal Infirmary either had elective surgery or underwent an emergency repair elsewhere. However, given the referral pathways that exist within this region (Leicester Royal Infirmary is the only vascular unit in Leicestershire) this is thought to be unlikely. 258 259 260 261 262 263 264 265 266 In conclusion, this study has shown that patients with a baseline AAA diameter of greater than 5.5cm and who were declined surgery or who did not undergo early elective surgical repair had improved survival rates compared with historical data. In particular, patients with 5.5-6.9cm AAA had a median survival of more than three years, suggesting that there is ample time for optimisation of risk factors and co-morbid pathologies before finally turning them down for surgery. By contrast, patients whose AAA exceeds 7cm face a very poor one year prognosis (35% rupture rate) and a more aggressive approach to management seems warranted.. 267 268 Conflict of Interest/Funding: None 269 270 271 1. Thompson MM. Controlling the expansion of abdominal aortic aneurysms. Br J Surg 2003;90:897 8 272 273 2. Lederle FA, Johnson GR, Wilson SE, et al. Rupture rate of large abdominal aortic aneurysms in patients refusing or unfit for elective repair. JAMA 2002;287:2968 72 15
274 275 276 3. Conway KP, Byrne J, Townsend M, Lane IF. Prognosis of patients turned down for conventional abdominal aortic aneurysm repair in the endovascular and sonographic era: Szilagyi revisited? Journal of Vascular Surgery 2001;33:752 7 277 278 4. Jones A, Cahill D, Gardham R. Outcome in patients with a large abdominal aortic aneurysm considered unfit for surgery. British Journal of Surgery 1998;85:1382 4 279 280 281 5. Western CE, Carlisle J, McCarthy RJ, Currie IC. Palliation of Abdominal Aortic Aneurysms in the Endovascular Era. European Journal of Vascular and Endovascular Surgery 2013;45:37 43 282 283 284 6. EVAR trial participants. Endovascular aneurysm repair and outcome in patients unfit for open repair of abdominal aortic aneurysm (EVAR trial 2): randomised controlled trial. Lancet 2005;365:2187 92 285 286 287 288 7. Buth J, van Marrewijk CJ, Harris PL, Hop WCJ, Riambau V, Laheij RJF. Outcome of endovascular abdominal aortic aneurysm repair in patients with conditions considered unfit for an open procedure: A report on the EUROSTAR experience. Journal of Vascular Surgery 2002;35:211 21 289 290 291 8. Sicard GA, Zwolak RM, Sidawy AN, White RA, Siami FS. Endovascular abdominal aortic aneurysm repair: Long-term outcome measures in patients at high-risk for open surgery. Journal of Vascular Surgery 2006;44:229 36 292 293 294 9. Thompson JP. Danjoux GR, Pichel A. Guidelines for the Provision of Anaesthetic Services; Chapter 15, vascular anaesthesia services 2015. Available at http://www.rcoa.ac.uk/gpas2015 16
295 296 297 298 10. Vascular Society of Great Britain and Ireland. Framework for Improving the Results of Elective AAA Repair [Internet]. 2009. Available from: http://www.vascularsociety.org.uk/library/quality-improvement/doc_download/25- framework-for-improving-the-results-of-elective-aaa-repair.html 299 300 301 302 303 11. Fleisher LA, Beckman JA, Brown KA, et al. ACC/AHA 2007 Guidelines on Perioperative Cardiovascular Evaluation and Care for Noncardiac Surgery A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 2002 Guidelines on Perioperative Cardiovascular Evaluation for Noncardiac Surgery). Circulation 2007;116:e418 e500 304 305 306 12. Karthikesalingam A, Holt PJ, Loftus IM, Thompson MM. Risk Aversion in Vascular Interventions: The Consequences of Publishing surgeon-specific mortality for Abdominal Aortic Aneurysm Repair. Eur J Vasc Endovasc Surg 2015 (IN PRESS) 307 308 13. Verzini F, de Rango P. Risk Aversion in Vascular Interventions. Eur J Vasc Endovasc Interv (IN PRESS) 17
Figure(s) Click here to download Figure(s): Unfit AAAs Figures (updated).ppt Figure 1: Outcomes for patients with AAA reviewed in the vascular surgical clinic. Surgical Clinic Review 764 Anaesthetic Clinic Review 575 Elective surgery 437 No Surgery 138 Mortality 71 (52%) Rupture 37 (52%) Non-rupture 34 (48%) Elective surgery 189
Figure 2: Kaplan-Meier analysis of all cause mortality within 36 months of clinic for the 138 patients with a documented AAA size not undergoing elective AAA repair. <7cm AAA 7cm AAA Time (months) 0 12 24 36 Numbers at risk AAA < 7cm 64 55 29 3 AAA 7cm 27 14 6 0
Figure 3: Kaplan-Meier analysis of mortality or urgent intervention within 36 months of clinic for 138 patients not undergoing elective AAA repair. <7cm AAA 7cm AAA Time (months) 0 12 24 36 Numbers at risk AAA < 7cm 64 55 29 3 AAA 7cm 27 14 6 0
Table(s) Click here to download Table(s): Unfit AAAs Tables (updated).ppt Table 1: Demographics for operated and non-operated patients Direct elective Non-operated p value surgery Gender: Male Female 393 (90%) 44 (10%) 123 (89%) NS 15 (11%) Age: Median (range) 73 (42-91) 76 (55-89) <0.001 Co-morbidities: Coronary artery disease 177 (41%) 60 (44%) NS Cardiac failure 16 (4%) 10 (7%) NS COPD/respiratory failure 91 (21%) 54 (39%) <0.001 Chronic kidney disease 45 (10%) 25 (18%) 0.01 Cerebrovascular disease 38 (9%) 24 (17%) 0.004 Diabetes 49 (11%) 12 (9%) NS Smoking status: Current or ex-smoker 107 (24%) 40 (29%) NS NS, not significant
Table 2: Aneurysm sizes in patients in non-operated group Male Female 5.5 5.9cm 42 (37%) 7 (30%) 6.0 6.9cm 47 (41%) 11 (48%) 7.0cm 26 (23%) 5 (22%)
Table 3: Reasons for patients not undergoing elective AAA surgery when advised to proceed at clinic Outcome following clinic Number (%) Comments On waiting list 14 (35%) Died from rupture event 3 Case 1: AAA size - 8.7cm Time to event - 2 months Case 2: AAA size - 8.5cm Time to event - 2 months Case 3: AAA size - 6cm Time to event - 29 months Emergency or urgent intervention 1 Case 1: AAA size - 5.7cm Time to event - 21 months Elective surgery after April 2012 6 Died from co-morbidity 3 Withdrawn from list due to health 1 Admitted from surveillance 1 (3%) Fit for EVAR but not anatomically appropriate 10 (25%) Patient subsequently declined surgery 8 (20%) Subsequently determined as unfit 6 (15%) Lost to follow-up 1 (3%) Total 40
Table 4: Cause of death for all patients that did not undergo elective surgery (*other causes of death include gastrointestinal bleed (2), old age (2), pulmonary embolism (1), chronic kidney disease (1)) Cause of death Number (%) Rupture 37 (52%) Ischaemic heart disease 9 (13%) Other* 6 (9%) Pneumonia 6 (8%) Cancer 4 (6%) Stroke 3 (4%) Dementia 3 (4%) COPD 2 (3%) Peripheral vascular disease 1 (1%) Total 71