Influence of Presentation on Outcome in Abdominal Aortic Aneurysm Repair

Eur J Vasc Endovasc Surg 32, 140–145 (2006) doi:10.1016/j.ejvs.2006.01.016, available online at http://www.sciencedirect.com on

Influence of Presentation on Outcome in Abdominal Aortic Aneurysm Repair D.A. Harris, A. Al-Allak, J. Thomas and A.R. Hedges* Princess of Wales Hospital, Coity Rd, Bridgend CF31 1RQ, UK Objectives. In the absence of formal screening abdominal aortic aneurysms (AAA) are detected in an opportunistic manner. Many remain asymptomatic and undetected until they rupture. Incidentally discovered small AAAs are entered into a surveillance programme until they reach a suitable size for repair. The aim of this study was to examine trends in the management of AAA and whether the method of presentation had an effect on subsequent mortality. Design. Observational study in UK district general hospital. Materials/methods. This study reports a single surgeon case series identified using a prospectively maintained database. Data on mode of presentation, management and mortality were retrieved from case notes, PIMS hospital database and the Office of National Statistics. Results. Two hundred and five patients were referred with AAAs between 1992 and 2004, 78% presenting in elective circumstances. The surveillance programme fed 33% of the operated cases. Two aneurysms ruptured whilst under surveillance. Overall elective operative mortality was 11.8% and has progressively decreased over time. Thirty-day operated mortality was significantly lower in patients having a period of surveillance than those having immediate elective repair (2.3 vs. 16.3%, pZ0.018). A slight reduction in emergency AAA repairs was noted over the study period (r2Z0.6) although registered aneurysm deaths continue to increase (r2Z0.83). Conclusions. Elective mortality following AAA surgery decreased over the study period. Outcome was better in those patients who had surgery for aneurysms that had been under surveillance. Despite opportunistic screening the population adjusted mortality rate of aortic aneurysms showed a progressive increase. A reduction in deaths from aneurysms is unlikely without a formal screening programme. Keywords: Aortic aneurysm; Surveillance; Mortality.

Introduction Aortic aneurysms are the 10th commonest cause of death in the UK with 8627 deaths in England and Wales in 2004.1 The incidence of ruptured abdominal aortic aneurysm (AAA) in Wales is eight per 100,000 population2 with an attendant risk of death of 40– 60%.3,4 The incidence of aortic aneurysms is consistently increasing in the UK.5 Abdominal aortic aneurysms (AAA) are traditionally detected in an opportunistic manner and often remain asymptomatic and undetected until they rupture.6,7 An increasing number of aortic aneurysms are being detected as a result of radiological investigation of unrelated conditions,8 and as a result are detected at a smaller size than that achieved by clinical

*Corresponding author. A.R. Hedges, MS, FRCS, Consultant Vascular Surgeon, Princess of Wales Hospital, Coity Rd, Bridgend CF31 1RQ, UK. E-mail address: [email protected]

examination. Currently, a policy of regular ultrasound surveillance until aneurysm size reaches 5.5 cm is employed and has been shown to be safe with an annual rupture rate of 1%. 9 Modern aneurysm management is concerned with the detection and timely elective repair of asymptomatic AAA thereby preventing rupture. Accordingly, there has been great interest in screening for this potentially life threatening condition. Certain patient groups are known to have an increased incidence of AAA such as those with hypertension, claudication10,11 and a family history of aneurysmal disease.12 Although selective screening of these groups has been considered, data from Gloucester and Chichester and more recently the MASS trial have demonstrated a significant reduction in mortality as a result of unselective screening programmes.13–15 It is clear that the modern detection of AAAs has been subject to much change of recent and will be radically different in the era of nationwide screening. It was, therefore, of interest to examine the current

1078–5884/000140 + 06 $35.00/0 q 2006 Elsevier Ltd. All rights reserved.

Mode of Presentation Influences AAA Mortality

demography of operated aneurysm cases and define the current practice of a vascular surgeon in a district general hospital prior to the introduction of a population screening programme.

Materials and Methods Cases were identified from a prospectively maintained database of aneurysm operations performed by a single surgeon (ARH). Non-operative cases were identified from hospital coding episodes, accident and emergency records and clinic letters. Aneurysms greater than 5.5 cm in males or 5 cm in females were considered suitable for repair. Detected aneurysms smaller than this were followed up in clinic by vascular nurse specialist using portable ultrasonography (Sonosite Inc., Bothell, WA). Screening intervals were yearly for aneurysms !4 cm, 6 monthly for those between 4 and 5 cm, and at 3 monthly intervals for O5 cm.16 Patients were considered suitable for aneurysm repair based on size of aneurysm, co-morbidity, left ventricular function, lung function testing and aneurysm characteristics based on CT scan using a specified aneurysm protocol. Whereas before 1992 elective and emergency aneurysm repairs were performed by general nonvascular surgeons, after 1992 all cases were dealt with by a vascular surgeon. Patient survival was calculated using mortality data obtained from the hospital Patient Information Management System (PIMS) database. Mortality statistics for the Bridgend area were obtained from data published by the Office for National Statistics. Disease codes 441 and I71 were used from the ninth and 10th revisions of the International Classification of Diseases (ICD-9)17 and ICD-10,18 respectively. Mortality rate was expressed per 100,000, with local population estimates obtained from the Office for National Statistics. Three patient groups are defined for the purposes of our study. Group 1 aneurysms were detected at less than 5.5 cm and had regular ultrasound surveillance until reaching the threshold for (delayed) elective repair. Group 2 aneurysms were detected at a size greater than 5.5 cm and had prompt elective repair without a surveillance period. Group 3 were emergency or urgent repair of ruptured or symptomatic aneurysms. The end points of the study were outcome of aneurysm detection (elective repair, emergency repair, ultrasonographic surveillance or conservative treatment) and 30-day operative mortality. We hypothesise

141

that incidental early aneurysm detection (group 1) is associated with a better outcome than those detected above the threshold for intervention (group 2). Statistical analysis was performed using SPSSw software version 11.5 (SPSS, Chicago, Illinois, USA). The chi-squared test was used for comparison of proportions and the Mann–Whitney test used for nonparametric comparison of continuous data. Kaplan– Meier survival curves were used to analyse differences in mortality, and the log-rank test used to determine differences. Cox regression analysis was used to detect association between mortality and confounding factors. Statistical significance was set at the 5% level.

Results Two hundred and five patients presented to a single consultant from 1992 to 2004 inclusive. One hundred and sixty-nine patients were operated upon by a single consultant (132 elective repairs and 37 emergencies). The remaining patients were either treated conservatively or declined intervention. One patient was referred for EVAR but was unsuitable. One third of the elective referrals were entered into a surveillance programme. Eighty-five percent (45/53) of these patients were eventually repaired, accounting for one third of all elective operations. The median interval between onset of aneurysm surveillance and surgery was 2 years (range 0.5–12 years). Two aneurysm ruptures were observed in patients undergoing surveillance (4.5 and 5.3 cm diameter). Of the asymptomatic referrals half were incidental radiological findings mostly through abdominal ultrasonography and computed tomography (85%). One third were incidental clinical findings, 10% were detected through ultrasound screening of claudicants referred to vascular clinic and the remainder were detected through sporadic screening of high risk groups. The number of elective procedures increased during the study period (Fig. 1) through an increasing number of operated surveillance cases (group 1, r2Z 0.999). In contrast, emergency AAA repairs decreased in the later years of the study. Thirty-day operative mortality for elective surgery (groups 1 and 2) showed a consistent decrease (rZ0.99) over the study period (Fig. 2). Thirty-day operated mortality was found to be significantly different when elective surgery is separated into immediate elective repair (group 2, 16.3%) vs. operating after a period of surveillance (group 1, 2.3%, pZ0.018). Survival benefits in the surveillance group persisted beyond 30 days. Five-year survival was 84% Eur J Vasc Endovasc Surg Vol 32, August 2006

D. A. Harris et al.

142

205 referrals 160 elective 89 repairs

53 surveillance

45 emergency 18 conservative

35 repairs

8 conservative

2 other

45 operated 43 elective 2 emergency 3 ongoing 3 later declined 1 discharged 1 died-refused

Fig. 1. Number of elective repairs and emergency repairs performed over the study period. Elective numbers are split into group 1 (operated surveillance cases) and group 2 (elective repair without surveillance).

for group 1, 57% for group 2 and 46% for group 3. Survival differences persisted until 8 years of follow up (Fig. 3). Aneurysm size, patient age and year of repair did not correlate with mortality differences based on Cox regression analysis (Table 1), although the surveillance group (1) was composed of more radiologically detected aneurysms than group 2 (78 vs. 48%, c2Z125, 4 df, p!0.0001). The population adjusted mortality rate for the catchment area served by the hospital is shown in Fig. 4, and shows a consistent increase in deaths from aortic aneurysms over the study period (r2Z0.83).

160 elective referrals

116 asymptomatic

10 symptomatic

58 radiological

43 clinical

12 claudicant screening

3 other

Fig. 2. Thirty-day operative mortality for elective AAA repair. Eur J Vasc Endovasc Surg Vol 32, August 2006

Discussion The aim of abdominal aortic aneurysm management is to treat the condition surgically before rupture intervenes due to continued growth. Accumulating evidence suggests that early detection of AAAs through population screening has a significant effect in reducing the mortality of this condition. Despite an initially high elective operated mortality a consistent decrease over time is observed. Our 12year overall mortality (11.8%) is higher than other published series 19 but represents the case mix presented. All cases have been submitted to the National Vascular Database with equal observed and predicted mortality. The consistent improvement in operative mortality is attributed to a combination of pre-operative cardiorespiratory optimisation and perioperative measures such as use of a dedicated vascular anaesthetist, selective pre-operative beta blockade and the introduction of cell salvage technology. Echocardiography assessment and selective lung function testing permits operative risk assessment and has improved case selection in line with current evidence.20 Perhaps the most important factor leading to reduced elective mortality is the increasing proportion of patients whose aneurysms had been detected early. Outcome was better in those patients who had surgery for small aneurysms that had been under surveillance (group 1) over those who did not have a surveillance period and underwent prompt repair (group 2).

Mode of Presentation Influences AAA Mortality

(i)

Elective repair

30

R2 = 0.25

Table 1. Multivariate analysis of risk factors for death after elective AAA repair

25 No surveillance period (group 2) Age Size of aneurysm Year of repair

20 15 10

143

Wald

Odds ratio

p*

4.9

2.2 (1.1–4.4)

0.026

3.2 0 7.4

1 (0.997–1.06) 1 (0.77–1.29) 0.87 (0.2–3.3)

0.07 0.99 0.68

Values in parentheses are 95% confidence intervals. * Cox proportional hazards model.

5 0 1992-1995

1996-1999

2000-2003

Year Emergency repair

14

R2 = 0.6048

12 10 8 6 4 2 0 1992-1995

1996-1999

2000-2003

Year (iii) 25

Operated surveillances

R2 = 0.9991

20 15 10 5 0 1992-1995

(iv) 16

1996-1999 Year Denied surgery

2000-2003

R2 = 0.9643

14 12 10 8

30

6 4 2 0 1992-1995

1996-1999

2000-2003

Year

30 day mortality (%)

(ii)

Differences in mortality were apparent at 30-day survival and persisted through 8 years of follow up. This phenomenon was not explained by patient age, aneurysm size or year of repair. Others21 have reported significantly lower 30-day mortality and long term survival after repair of screen-detected aneurysms compared to incidentally discovered aneurysm. This difference was, however, attributed to a lower age of the screened population, which does not explain our findings. We propose two explanations for the difference in mortality between the two groups. Firstly, if aneurysms are detected early there is opportunity to control cardiovascular risk factors. All such patients are commenced on aspirin, ACE inhibitors and a statin; hypertension is aggressively controlled and smoking cessation instituted. For those aneurysms requiring immediate repair there is no such opportunity for a period of optimisation. It could also be argued that those who adhere to a surveillance programme are more proactive about their health and are more likely to comply with risk factor modification. Secondly, the surveillance group may have lower mortality as those unfit for surgery were excluded at presentation. Those presenting with larger aneurysms are more likely to be advised to have aneurysm repair because of the higher rupture risk, despite having potentially worse co-morbidity. It is easier to turn down an unfit patient with a small AAA than an unfit Elective 30 day operated mortality

25 20 15

R2 = 0.9991

10 5 0

Fig. 3. Kaplan–Meier survival curves for group 1 (operated surveillance), group 2 (timely elective repair without surveillance) and group 3 (emergency surgery). Difference between operated surveillance cases and immediate elective repair pZ0.02, log-rank test.

1992-1995

1996-1999

2000-2003

Year

Fig. 4. Registered aortic aneurysm mortality from 1993 to 2003 for the Bridgend area per 100,000 population. Eur J Vasc Endovasc Surg Vol 32, August 2006

144

D. A. Harris et al.

one with a larger aneurysm at higher risk of rupture. Data from Gloucestershire22 suggests that patients with screen detected aneurysms have more favourable pre-operative physiology scores as assessed by P-POSSUM which would fit with this latter hypothesis. The majority of patients in our surveillance programme had repair of their aneurysm (85%) in contrast to other reports where 62% were repaired.23 This may reflect our policy of not performing surveillance on those patients unfit for surgery and a longer follow up period. We noted two patients that had ruptures of small aneurysms (mean size 4.9 cm) whilst under surveillance (0.3% p.a.). This rate compares favourably with other studies who quote rupture rates of 0.6–1% p.a.9,23 Despite opportunistic screening and maintaining an aneurysm surveillance programme with an attendant decrease in emergency aortic surgery, the population adjusted mortality rate of aortic aneurysms in the Bridgend area showed a progressive increase over the period studied. This mirrors the continuing escalation of death rates caused by AAAs in England and Wales24 that may be explained by a true increase in aneurysm incidence coupled with a failure to reduce deaths from rupture of undetected aneurysms. This leads back to the argument that mortality from aortic aneurysms will only decline in the presence of a population screening programme.15 Although screening of males over 65 years will reduce AAA mortality it will not make emergency repair a redundant procedure. In our series, 35 aneurysms were detected in the under 65-age group which includes six ruptures (15% of all ruptured AAAs). Although screening proposals exclude females based on findings of similar incidences of AAA rupture in screened and control groups25 we found that 15% of our aneurysm referrals were female and one third of these presented as emergencies. We would argue that such recommendations do not take into account regional variations and that AAA screening indications may be extended in some areas. Despite a recommendation by the US Preventive Services Task Force26 to screen men aged 65–75 for AAA and a growing body of evidence demonstrating reduction in AAA mortality as a result of screening, the UK National Screening Committee continue to keep plans for a population screening programme under review in spite of evidence proving its cost effectiveness.27 Our paper adds weight to the argument for early detection of asymptomatic aneurysms as these patients, who have aneurysm repair when the aortic diameter reached a pre-determined size, have Eur J Vasc Endovasc Surg Vol 32, August 2006

both an immediate and long lasting protective effect against post-repair mortality.

References 1 Source www.statistics.gov.uk/STATBASE/Expodata/Spreadsheets/D8986.xls. 2 Basnyat PS, Biffin AH, Moseley LG, Hedges AR, Lewis MH. Mortality from ruptured abdominal aortic aneurysm in Wales. Br J Surg 1999;86:765–770. 3 Bown MJ, Sutton AJ, Bell PR, Sayers RD. A meta-analysis of 50 years of ruptured abdominal aortic aneurysm repair. Br J Surg 2002;89:714–730. 4 Bradbury AW, Adam DJ, Makhdoomi KR, Stuart WP, Murie JA, Jenkins AM et al. A 21-year experience of abdominal aortic aneurysm operations in Edinburgh. Br J Surg 1998;85:645– 647. 5 Fowkes FG, Macintyre CC, Ruckley CV. Increasing incidence of aortic aneurysms in England and Wales. BMJ 1989;298:33–35. 6 Fielding JW, Black J, Ashton F, Slaney G, Campbell DJ. Diagnosis and management of 528 abdominal aortic aneurysms. BMJ 1981;283:355–359. 7 Craig SR, Wilson RG, Walker AJ, Murie JA. Abdominal aortic aneurysm: still missing the message. Br J Surg 1993;80:450–452. 8 Davies AJ, Winter RK, Lewis MH. Prevalence of abdominal aortic aneurysms in urology patients referred for ultrasound. Ann R Coll Surg Engl 1999;81:235–238. 9 The UK Small Aneurysm Trial Participants. Mortality results for randomised controlled trial of early elective surgery or ultrasonographic surveillance for small abdominal aortic aneurysms. Lancet 1998;352:1649–1655. 10 Galland RB, Simmons MJ, Torrie EP. Prevalence of abdominal aortic aneurysm in patients with occlusive peripheral vascular disease. Br J Surg 1991;78:1259–1260. 11 Macsweeney ST, O’Meara M, Alexander C, O’Malley MK, Powell JT, Greenhalgh RM. High prevalence of unsuspected abdominal aortic aneurysm in patients with confirmed symptomatic peripheral or cerebral arterial disease. Br J Surg 1993; 80:582–584. 12 Bengtsson H, Sonesson B, Bergqvist D. Incidence and prevalence of abdominal aortic aneurysms, estimated by necropsy studies and population screening by ultrasound. Ann N Y Acad Sci 1996;800:1–24. 13 Heather BP, Poskitt KR, Earnshaw JJ, Whyman M, Shaw E. Population screening reduces mortality rate from aortic aneurysm in men. Br J Surg 2000;87:750–753. 14 Scott RA, Wilson NM, Ashton HA, Kay DN. Influence of screening on the incidence of ruptured abdominal aortic aneurysm: 5-year results of a randomized controlled study. Br J Surg 1995;82:1066–1070. 15 Ashton HA, Buxton MJ, Day NE, Kim LG, Marteau TM, Scott RA et al. The Multicentre Aneurysm Screening Study (MASS) into the effect of abdominal aortic aneurysm screening on mortality in men: a randomised controlled trial. Lancet 2002; 360:1531–1539. 16 Cook TA, Galland RB. A prospective study to define the optimum rescreening interval for small abdominal aortic aneurysm. Cardiovasc Surg 1996;4:441–444. 17 Manual of the statistical international classification of diseases, injuries and causes of death. 1975 Revision. Geneva, World Health Organisation; 1977. 18 International statistical classification of diseases and related health problems. 1989 Revision. Geneva, World Health Organisation; 1992. 19 Bayly PJ, Matthews JN, Dobson PM, Price ML, Thomas DG. Inhospital mortality from abdominal aortic surgery in Great Britain and Ireland: Vadcular Anaesthesia Society audit. Br J Surg 2001; 88:687–692.

Mode of Presentation Influences AAA Mortality

20 Brady AR, Fowkes FG, Greenhalgh RM, Powell JT, Ruckley CV, Thompson SG. Risk factors for postoperative death following elective surgical repair of abdominal aortic aneurysm: results from the UK Small Aneurysm Trial. On behalf of the UK Small Aneurysm Trial participants. Br J Surg 2000; 87:742–749. 21 Taylor JC, Shaw E, Whyman MR, Poskitt KR, Heather BP, Earnshaw JJ. Late survival after elective repair of aortic aneurysms detected by screening. Eur J Vasc Endovasc Surg 2004;28:270–273. doi:10.1016/j.ejvs.2004.05.007. 22 Irvine CD, Shaw E, Poskitt KR, Whyman MR, Earnshaw JJ, Heather BP. A comparison of the mortality rate after elective repair of aortic aneurysms detected either by screening or incidentally. Eur J Vasc Endovasc Surg 2000;20(374):378. doi:10. 1053/ejvs.2000.1187. 23 Lederle FA, Wilson SE, Johnson GR, Reinke DB, Littooy FN, Acher CW et al. Immediate repair compared with surveillance of small abdominal aortic aneurysms. N Engl J Med 2002; 346:1437–1444.

145

24 Filipovic M, Goldacre MJ, Roberts SE, Yeates D, Duncan ME, Cook-Mozaffari P. Trends in mortality and hospital admission rates for abdominal aortic aneurysm in England and Wales, 1979–1999. Br J Surg 2005;92:968–975. 25 Scott RA, Bridgewater SG, Ashton HA. Randomized clinical trial of screening for abdominal aortic aneurysm in women. Br J Surg 2002;89:283–285. 26 US Services Task Force Preventive. Screening for abdominal aortic aneurysm: recommendation statement. Ann Int Med 2005; 142:198–202. 27 Multicentre Aneurysm Screening Study Group. Multicentre Aneurysm Screening Study (MASS): cost effectiveness analysis of screening for abdominal aortic aneurysms based on four year results from randomised controlled trial. BMJ 2002; 325:1135. Accepted 25 January 2006 Available online 3 April 2006

Eur J Vasc Endovasc Surg Vol 32, August 2006