Volume 39, Issue 6 , Pages 719-725, June 2010
Anaesthetic Specialisation Leads to Improved Early- and Medium-term Survival Following Major Vascular Surgery
Article Outline
Abstract
Objective
Vascular surgical specialisation is associated with improved outcomes. We aimed to assess the effect of anaesthetic specialisation on outcome following major vascular surgery.
Design
Retrospective cohort study.
Methods
Patients undergoing major vascular surgery (lower limb revascularisation, elective and ruptured abdominal aortic aneurysm repair, endovascular aneurysm repair and carotid endarterectomy) over a five-year period were identified from a prospective database. The primary outcomes were death within 30 days and death within two years of surgery. Potential risk factors for mortality were assessed using multivariate logistic regression modelling.
Results
The analysis cohort comprised 1155 patients followed up for a median of 583 days. Mortality within two years of surgery was 16%. For the overall cohort, care from vascular anaesthetists was independently associated with reduced 30-day (odds ratio 0.22; 95% CI 0.12–0.62) and medium-term mortality (0.31; 95% CI 0.18–0.55). For elective patients (n
=851), vascular anaesthesia reduced two-year mortality (odds ratio 0.29; 95% CI 0.15–0.58; P=0.0004) though not 30-day mortality (odds ratio 0.55; 95% CI 0.15–1.95; P=0.35). For emergency patients, care by a vascular anaesthetist influenced neither 30-day mortality (odds ratio 0.33; 95% CI 0.08–1.41; P=0.13) nor medium-term mortality (odds ratio 0.45; 95% CI 0.17–1.21; P=0.11).
Conclusions
Anaesthetic specialisation reduced early- and medium-term mortality rates following major vascular surgery. If replicated by prospective studies, these results suggest that vascular surgery services would benefit from specialised anaesthetic support.
Keywords: Anesthesia/mortality, Surgical procedures, Operative/mortality
Introduction
Surgical specialisation and high-volume practice both improve outcome following major complex surgery.1, 2 Similar patterns exist in a range of vascular procedures, including carotid endarterectomy,3 abdominal aortic aneurysm (AAA) repair 4 and lower limb revascularisation.5 Peri-operative mortality from major vascular surgery has decreased slightly in recent years, partly through increased use of endovascular techniques.6 Nevertheless, immediate and medium-term mortality remain considerable.7, 8
In contrast to surgery, anaesthesia has largely remained a broad-based, general speciality. Contemporary data from the United Kingdom reveal that one-fifth of elective and over half of ruptured AAA patients are cared for by anaesthetists who undertake less than five AAA repairs per annum.9 Emerging surgical oncology data suggest that peri-operative anaesthetic techniques may have an effect on eventual oncological outcome.10 Little is known about the relationship between anaesthetic subspecialisation and outcome following major vascular surgery. We hypothesised that care provided by specialist vascular anaesthetists would improve peri-operative and medium-term outcome following major vascular surgery.
Methods
The Norfolk and Norwich Vascular Unit is a tertiary referral vascular unit serving a population of approximately eight hundred thousand people. The unit undertakes a full range of emergency and elective vascular surgery. All elective and emergency vascular surgeries are provided by five consultant vascular surgeons. There are six consultant anaesthetists with a special interest in vascular surgery. These anaesthetists undertake at least three elective vascular surgery lists each month and are members of the United Kingdom Vascular Anaesthesia Society. There is no separate vascular anaesthesia on-call rota. Emergency vascular patients are cared for by the anaesthetist of the day, who may or may not have a vascular interest.
All patients undergoing surgery in the unit are entered into a prospective database, based upon the Vascular Society of Great Britain and Ireland's National Vascular Database. Age, pre-operative blood results, co-morbidities (diabetes, smoking, hypertension, ischaemic heart disease, renal failure), electrocardiogram findings, operating surgeon, attending anaesthetist, type of surgery, blood loss, duration of surgery, elective versus emergency procedure, intra-operative haemodynamic variables, post-operative complications and death within 30 days of surgery are recorded for each patient.
Patients who underwent surgery between 2003 and 2007 were extracted from this database. The patient list was then cross-referenced with the monthly departmental audit records and surgeon operating logs to ensure complete data capture. Only patients undergoing an indexed procedure (elective open AAA repair, rupture AAA repair, endovascular AAA repair, carotid endarterectomy, lower limb revascularisation) were included in the analyses. Those patients undergoing other procedures e.g. haemodialysis access formation, major or minor limb amputation or vascular trauma were excluded. Patients undergoing venous surgery are not entered into the database.
Logistic regression models were constructed to identify factors correlating with 30-day mortality and medium-term survival. Medium-term survival was defined as survival up to two years post-operatively. Patients were censored when they died, were lost to follow-up or exceeded two years follow-up. Medium-term survival was determined by cross-referencing the patient cohort with the hospital administrative database. For patients who underwent multiple procedures, only the first procedure was included in the analyses. Univariate logistic regression models were used to test the effect of each potential prognostic variable on outcome. Each model initially considered the following variables: haemoglobin level, white cell count, serum sodium, serum potassium, serum urea, serum creatinine, age, male gender, mode of admission (elective or emergency), diabetes, ischaemic heart disease or congestive cardiac failure, hypertension, smoking status, abnormal pre-operative electrocardiogram, lowest intra-operative systolic blood pressure, highest intra-operative heart rate, operative time, intra-operative blood loss (<1L, 1–2L, 2–5L or >5L). The type of surgery was also included in each model (lower limb revascularisation, elective open abdominal aortic aneurysm, elective endovascular aneurysm repair, carotid endarterectomy, open repair of ruptured abdominal aortic aneurysm). Lower limb revascularisation was used as the reference group to which the other procedure types were compared.
Significant univariate correlates (defined as P≤0.1) were then entered into forward stepwise logistic regression models to test for independence. A p-value of 0.05 or less was considered significant in the multivariate models. Models were constructed for the overall patient cohort. Further models were constructed using datasets restricted to elective patients only, emergency patients only and with ruptured AAA patients excluded. Factors influencing survival were assessed by Kaplan–Meier and Cox regression as required. All p-values are two-sided. Continuous variables are reported as mean (standard deviation) and categorical variables as absolute number (%), unless stated otherwise. The statistical analyses were performed using Statsdirect 2.7.3 (Statsdirect Ltd, Altrincham, UK).
Results
The initial search yielded 2295 patients. Of these 1155 had undergone an indexed procedure. The majority of the cohort were male, three-quarters underwent elective procedures and about one-quarter were cared for by a vascular anaesthetist (Table 1). Candidate variables which did not reach significance in the univariate analyses have been omitted from the results tables.
Table 1. Characteristics of patients eligible for inclusion.
| Characteristic | Overall cohort (n=1155) | Elective only (n=851) | Emergency only (n=304) |
|---|---|---|---|
| Mean age±SD | 72±8.5 years | 71±8.5 years | 74±8.4 years |
| Male gender | 858 (74%) | 618 (73%) | 240 (79%) |
| Type of procedure | |||
| Elective AAA repair | 306 (26%) | 305 (36%) | – |
| Rupture AAA repair | 206 (18%) | – | 206 (68%) |
| EVAR | 56 (5%) | 56 (7%) | 1 (0.3%) |
| Carotid endarterectomy | 347 (30%) | 317 (37%) | 39 (12.7%) |
| Lower limb revascularisation | 240 (21%) | 174 (20%) | 58 (19%) |
| Peri-operative deaths | 63 (6%) | 15 (2%) | 48 (16%) |
| Deaths within 2 years | 179 (16%) | 95 (11%) | 84 (28%) |
| Vascular anaesthetist present | 297 (26%) | 265 (31%) | 32 (11%) |
Peri-operative mortality
Factors affecting 30-day mortality were initially modelled using the entire cohort of 1155 patients. Univariate logistic regression modelling identified a number of factors that influenced peri-operative mortality in this cohort (Table 2). In the multivariate model, care by a vascular anaesthetist reduced mortality (odds ratio 0.33; 95% CI 0.12–0.62; P=0.002). Pre-operative serum creatinine had a small adverse influence on survival (odds ratio 1.00; 95% CI 1.00–1.01). Surgery for a ruptured AAA, peri-operative blood loss and increasing age were all associated with an increased risk of peri-operative death. Given the strong influence of ruptured AAA and intra-operative blood loss on 30-day mortality, the analysis was repeated with all emergency patients excluded. The elective only model encompassed 851 patients, of whom 15 died peri-operatively (2%). In this model, care by a vascular anaesthetist had no influence on outcome (odds ratio 0.55; 95% CI 0.15–1.95) (Table 3). Similarly, care by a vascular anaesthetist did not affect 30-day outcome in patients undergoing emergency vascular surgery (Table 4, Table 5, Table 6, Table 7).
Table 2. Factors affecting 30-day survival in the entire cohort (1155 patients).
| Variable | Alive at 30 days (n=1092) | Death within 30 days (n=63) | Univariate odds ratio (95% CI) | P | Multivariate odds ratio (95% CI) | p |
|---|---|---|---|---|---|---|
| Vascular anaesthetist | 292 (27%) | 5 (8%) | 0.24 (0.09–0.59) | 0.002 | 0.22 (0.12–0.62) | 0.002 |
| Creatinine (μmol/L) | 108 (44.4) | 146 (64.7) | 1.01 (1.00–1.02) | <0.0001 | 1.01 (1.00–1.01) | 0.02 |
| Age (years) | 72 (8.5) | 77 (6.0) | 1.09 (1.06–1.13) | <0.0001 | 1.08 (1.03–1.14) | 0.0008 |
| Blood loss | ||||||
| >5L | 6 (0.5%) | 6 (9%) | 91.8 (21.87–385.2) | <0.0001 | 18.12 (3.58–91.5) | 0.005 |
| 1–2L | 281 (26%) | 36 (57%) | 11.76 (4.56–30.32) | <0.0001 | 5.10 (1.83–14.27) | 0.002 |
| 2–5L | 33 (3%) | 8 (13%) | 22.25 (6.89–71.84) | <0.0001 | 4.17 (1.05–16.47) | 0.04 |
| Procedure | ||||||
| Open repair of ruptured AAA | 159 (15%) | 47 (75%) | 70.65 (9.65–517.2) | <0.0001 | 5.47 (2.62–11.46) | <0.001 |
| EVAR | 54 (5%) | 2 (3%) | 8.85 (0.79–99.40) | 0.07 | 2.62(0.50–13.69) | 0.25 |
| Elective open AAA repair | 295 (27%) | 11 (18%) | 8.91 (1.14–69.52) | 0.04 | 0.75 (0.38–36.66) | – |
| 0.26 | ||||||
| Haemoglobin (g/L) | 13.4 (3.43) | 11.3 (2.45) | 0.68 (0.61–0.76) | <0.0001 | 0.92 (0.79–1.07) | 0.28 |
| White cell count (×109/L) | 9.2 (3.86) | 11.9 (5.41) | 1.10 (1.05–1.15) | <0.0001 | 1.04 (0.96–1.13) | 0.31 |
| Elective admission | 835 (77%) | 15 (24%) | 0.10 (0.05–0.17) | <0.0001 | 127,387 (0.01–150,000) | 0.98 |
| Abnormal ECG | 485 (44%) | 36 (57%) | 2.76 (1.39–5.50) | 0.003 | 1.66 (0.77–3.59) | 0.19 |
| Lowest intra-operative systolic blood pressure (mmHg) | 126 (33.4) | 106 (40.2) | 0.98 (0.97–0.99) | 0.0003 | 0.99 (0.98–1.01) | 0.99 |
| Highest intra-operative heart rate (beats per minute) | 78 (16.8) | 93 (21.9) | 1.04 (1.03–1.05) | <0.0001 | 1.01 (0.99–1.03) | 0.46 |
Table 3. Factors affecting 30-day mortality in elective patients (n=851).
| Variable | Alive at 30 days (n=836) | Dead at 30 days (n=15) | Univariate OR | P | Multivariate OR | p |
|---|---|---|---|---|---|---|
| Vascular anaesthetist | 262 (31%) | 3 (20%) | 0.55 (0.15–1.95) | 0.35 | – | – |
| Age (years) | 71.6 (8.49) | 75.9 (5.12) | 1.07 (0.99–1.15) | 0.05 | 1.07 (0.98–1.16) | 0.12 |
| Operation time (min) | 161 (54.5) | 207 (108.2) | 1.01 (1.00–1.02) | 0.005 | 1.01 (1.00–1.02) | 0.008 |
| Procedure | ||||||
| Carotid endarterectomy | 316 (38%) | 1 (7%) | 0.12 (0.12–0.90) | 0.04 | 3.67 (0.71–18.95) | 0.68 |
| Elective open AAA repair | 294 (35%) | 11 (73%) | 5.07 (1.60–16.06) | 0.006 | 4.77 (1.28–14.77) | 0.02 |
Table 4. Factors affecting 30-day mortality in emergency patients (n=304).
| Variable | Alive at 30 days (n=256) | Death within 30 days (n=48) | Univariate OR | P | Multivariate OR | p |
|---|---|---|---|---|---|---|
| Vascular anaesthetist | 30 (12%) | 2 (4%) | 0.33 (0.08–1.41) | 0.13 | – | – |
| Creatininea | 104 (88–131) | 151 (112–182) | 1.02 (1.01–1.02) | <0.0001 | 1.01 (1.00–1.02) | 0.0003 |
| Age (years) | 73 (8.7) | 78 (6.3) | 1.08 (1.03–1.14) | 0.0003 | 1.11 (1.05–1.18) | 0.0004 |
| Blood loss | ||||||
| >5L | 6 (2%) | 5 (10%) | 4.23 (1.25–14.83) | 0.02 | 12.26 (2.47–60.65) | 0.002 |
| 1–2L | 93 (36%) | 30 (63%) | 2.80 (1.35–5.77) | 0.006 | 6.43 (2.29–18.05) | 0.0004 |
| Haemoglobin (g/dL) | 12.6 (2.57) | 10.7 (2.19) | 0.75 (0.66–0.86) | <0.0001 | 0.90 (0.77–1.06) | 0.21 |
| Abnormal ECG | 121 (47%) | 27 (56%) | 2.74 (1.14–6.58) | 0.02 | 1.73 (0.62–4.81) | 0.29 |
| Lowest intra-operative systolic blood pressure (mmHg) | 110 (37.4) | 96 (38.3) | 0.99 (0.98–1.00) | 0.05 | 0.99 (0.98–1.01) | 0.73 |
| Highest heart rate (beats per minute) | 88 (19.8) | 98 (22.9) | 1.02 (1.01–1.04) | 0.008 | 1.01 (0.99–1.04) | 0.25 |
| Procedure | ||||||
| Rupture AAA | 159 (62%) | 47 (98%) | 28.67 (3.89–211.17) | 0.001 | 1,735,307 (0.01 to ∞) | 0.98 |
| Carotid | 29 (11%) | 1 (2%) | 0.17 (0.02–1.25) | 0.08 | 0.96 (0.01–2.00) | 0.99 |
aMedian (interquartile range); ECG: electrocardiogram; AAA; abdominal aortic aneurysm. |
Table 5. Logistic regression analyses of factors affecting medium-term survival in the entire cohort (1155 patients).
| Variable | Alive at 2 years (n=976) | Death within 2 years (n=179) | Univariate OR | P | Multivariate OR | p |
|---|---|---|---|---|---|---|
| Vascular anaesthetist | 280 (29%) | 20 (11%) | 0.30 (0.19–0.49) | <0.0001 | 0.31 (0.18–0.55) | <0.0001 |
| Age (years) | 71 (8.5) | 76 (6.6) | 1.11 (1.08–1.14) | <0.0001 | 1.10 (1.06–1.12) | <0.0001 |
| Haemoglobin (g/dL) | 13.5 (3.54) | 12.1 (2.33) | 0.74 (0.69–0.80) | <0.0001 | 0.86 (0.79–0.94) | 0.001 |
| White cell count (x109/L) | 9.2 (3.91) | 10.8 (4.76) | 1.08 (1.05–1.12) | <0.0001 | 1.06 (1.02–1.10) | 0.004 |
| Creatinine (μmol/L)a | 99 (84–119) | 117 (93–151) | 1.01 (1.00–1.01) | <0.0001 | 1.01 (1.00–1.02) | 0.01 |
| Urea (mg/L)a | 6.3 (5.1–8.0) | 7.8 (5.7–10.1) | 1.01 (0.99–1.02) | 0.07 | 0.96 (0.88–1.04) | 0.32 |
| Sodium (mmol/L) | 139 (12.7) | 137 (14.8) | 0.99 (0.97–0.99) | 0.02 | 0.99 (0.96–1.02) | 0.39 |
| Elective admission | 756 (78%) | 96 (54%) | 0.31 (0.23–0.43) | <0.0001 | 0.96 (0.53–1.76) | 0.91 |
| Ischaemic heart disease/congestive cardiac failure | 361 (37%) | 74 (41%) | 1.40 (0.98–1.98) | 0.06 | 1.23 (0.78–1.93) | 0.38 |
| Abnormal ECG | 437 (45%) | 84 (47%) | 1.49 (1.02–2.19) | 0.04 | 0.83 (0.47–1.47) | 0.53 |
| Highest heart rate (beats per minute) | 78 (16.5) | 86 (21.6) | 1.02 (1.01–1.04) | <0.0001 | 1.01 (0.99–1.02) | 0.29 |
| Blood loss | ||||||
| >5L | 8 (1%) | 4 (2%) | 16.94 (4.92–58.31) | <0.0001 | 3.45 (0.89–13.33) | 0.07 |
| 1–2L | 253 (26%) | 64 (36%) | 2.14 (1.43–3.21) | 0.0002 | 1.49 (0.96–2.31) | 0.07 |
| 2–5L | 25 (3%) | 16 (9%) | 5.42 (2.71–10.85) | <0.0001 | 2.01 (0.91–4.47) | 0.08 |
| Procedure | ||||||
| Open repair of ruptured AAA | 136 (14%) | 70 (39%) | 2.92 (1.85–4.61) | <0.0001 | 0.82 (0.28–2.56) | 0.76 |
| Carotid endarterectomy | 315 (32%) | 32 (18%) | 0.58 (0.35–0.96) | 0.03 | 1.12 (0.53–1.29) | 0.76 |
aMedian (interquartile range) ECG: electrocardiogram; AAA: abdominal aortic aneurysm. |
Table 6. Factors affecting medium-term survival in elective patients (n=851).
| Variable | Alive at 2 years (n=756) | Death within 2 years (n=95) | Univariate OR | P | Multivariate OR | p |
|---|---|---|---|---|---|---|
| Vascular anaesthetist | 251 (33%) | 14 (15%) | 0.35 (0.19–0.62) | 0.0004 | 0.29 (0.15–0.58) | 0.0004 |
| Age (years) | 71 (8.5) | 76 (6.3) | 1.09 (1.05–1.13) | <0.0001 | 1.09 (1.05–1.13) | <0.0001 |
| Haemoglobin (g/dL) | 13.7 (3.75) | 12.9 (1.93) | 0.81 (0.72–0.91) | 0.0004 | 0.96 (0.81–1.13) | 0.63 |
| Creatinine (μmol/L) | 106 (41.5) | 124 (67.2) | 1.01 (1.00–1.01) | 0.001 | 1.00 (0.99–1.01) | 0.36 |
| Lowest intra-operative systolic blood pressure (mmHg) | 129 (30.8) | 140 (27.6) | 1.01 (1.00–1.02) | 0.008 | 1.01 (0.99–1.02) | 0.10 |
| Procedure | ||||||
| Carotid endartectomy | 289 (38%) | 28 (29%) | 0.55 (0.31–0.97) | 0.04 | 0.98 (0.58–1.69) | 0.96 |
Table 7. Factors affecting medium-term survival in emergency patients (n=304).
| Variable | Alive at 2 years (n=220) | Death within 2 years (n=84) | Univariate OR | P | Multivariate OR | p |
|---|---|---|---|---|---|---|
| Vascular anaesthetist | 27 | 5 | 0.45 (0.17–1.21) | 0.11 | – | – |
| Age (years) | 72 (8.5) | 79 (6.7) | 1.12 (1.07–1.16) | <0.0001 | 1.15 (1.09–1.22) | <0.0001 |
| Creatinine (μmol/L)a | 103 (88–124) | 132 (96–167) | 1.01 (1.00–1.02) | <0.0001 | 1.01 (1.00–1.02) | 0.01 |
| Blood loss | ||||||
| >5L | 4 (2%) | 7 (8%) | 15.30 (3.67–63.97) | 0.0002 | 8.38 (1.65–42.66) | 0.01 |
| 1–2L | 84 (38%) | 39 (46%) | 4.06 (1.78–9.26) | 0.0009 | 3.90 (1.50–10.12) | 0.005 |
| 2–5L | 13 (6%) | 14 (17%) | 9.42 (3.29–26.96) | <0.0001 | 6.41 (1.80–22.81) | 0.004 |
| Sodium (mmol/L) | 139 (3.4) | 136 (15.6) | 0.91 (0.85–0.97) | 0.004 | 0.90 (0.82–1.00) | 0.05 |
| Haemoglobin (g/dL) | 12.7 (2.55) | 11.2 (2.43) | 0.79 (0.71–0.88) | <0.0001 | 0.94 (0.81–1.09) | 0.43 |
| Ischaemic heart disease/congestive cardiac failure | 75 | 39 | 1.81 (1.05–3.13) | 0.03 | 1.62 (0.80–3.27) | 0.18 |
| Highest heart rate (beats per minute) | 87 (19.6) | 98 (22.2) | 1.02 (1.01–1.04) | 0.004 | 0.99 (0.98–1.02) | 0.77 |
| Procedure | ||||||
| Open repair of ruptured AAA | 136 (62%) | 70 (83%) | 3.60 (1.21–10.68) | 0.02 | 0.54 (0.09–2.97) | 0.48 |
Medium-term outcome
Median follow-up for the cohort was 583 days. There were 179 deaths during follow-up (16%). Mortality for the subgroup cared for by vascular anaesthetists was 7% (20 deaths in 287 patients). For the overall cohort, care by a vascular anaesthetist reduced the risk of death within two years (Fig. 1) (hazard ratio 0.37; 95% CI 0.26–0.51; P<0.0001). The benefit persisted when the analysis was restricted to elective patients only (hazard ratio 0.44; 95% CI 0.28–0.69; P=0.004) (Fig. 2). The survival difference did not reach statistical significance when the analysis was restricted to emergency patients only (hazard ratio 0.48; 95% CI 0.25–0.93; P=0.09) (Fig. 3). Univariate logistic regression again identified a number of significant correlates with survival, including vascular anaesthesia (Table 5). In a multivariate model, vascular anaesthesia was independently associated with a reduction in mortality (multivariate odds ratio 0.31; 95% CI 0.18–0.55). Although the protection afforded by vascular anaesthesia appeared independent of a diagnosis of rupture AAA in this model, blood loss and pre-operative haemoglobin level were associated with an increased mortality risk. To account for potential confounding by the inclusion of ruptured aneurysm and other emergency patients, we repeated the model using elective patients only (Table 6). In this model (851 patients; 95 deaths) the only univariate correlates with survival were vascular anaesthesia (odds ratio 0.29; 95% CI 0.15–0.58), age (odds ratio 1.09; 95% CI 1.05–1.13) and intra-operative systolic blood pressure (odds ratio 1.01; 95% CI 0.99–1.02). In the multivariate model, only vascular anaesthesia retained significance (multivariate odds ratio 0.44; 95% CI 0.23–0.79; P=0.004).
Figure 1
Survival curves for 1155 patients receiving care from a general anaesthetist (○; Group 0) or a vascular anaesthetist (□; Group 1). Numbers at risk (standard error of survival hazard) are displayed along the horizontal axis.
Figure 2
Survival curves for 851 elective patients receiving care from a general anaesthetist (○; Group 0) or a vascular anaesthetist (□; Group 1). Numbers at risk (standard error of survival hazard) are displayed along the horizontal axis.
Figure 3
Survival curves for 304 emergency patients receiving care from a general anaesthetist (○; Group 0) or a vascular anaesthetist (□; Group 1). Numbers at risk (standard error of survival hazard) are displayed along the horizontal axis.
The models were repeated using datasets restricted to emergency patients only, and excluding ruptured AAA patients only (Table 7). Care by vascular anaesthetists did not correlate with survival in these further models.
Discussion
Recent years have seen a rapid increase in surgical sub-specialisation, driven by recognition of the relationship between volume, specialisation and surgical outcome. In vascular surgery, most procedures are now performed by vascular surgeons and trainees, rather than by general surgeons.11 In contrast, anaesthesia remains broad-based. Apart from intensive care, paediatric and cardiac anaesthesia, the majority of anaesthetists remain generalists, taking care of patients undergoing a variety of surgical procedures. The present data suggest that a degree of anaesthetic specialisation is associated with reductions in both peri-operative and medium-term mortality.
In models using the entire cohort of 1155 patients, vascular anaesthesia significantly reduced both 30-day and medium-term mortality. When the analysis was restricted to elective patients only, vascular anaesthesia retained benefit with respect to medium-term outcome. There was no effect on 30-day outcome. However, with only 15 peri-operative deaths in the elective group, the analysis is likely to be underpowered. Our results should not be interpreted as ruling out benefit in this subgroup. Similarly, the relatively small size of the emergency subgroup (304 patients) means that benefit cannot be ruled out in these patients. The survival curves for emergency patients are diverging at two years (Fig. 3) and the 95% confidence interval for the hazard ratio does not cross one. Further follow-up or further patients may yield a significant difference.
Why should vascular anaesthetic specialisation improve medium-term outcome, long after the peri-operative period? Asymptomatic peri-operative cardiac injury is associated with increased mortality for several years after major vascular surgery.12, 13 Peri-operative haemodynamic stability may reduce the risk of peri-operative myocardial injury so one can speculate that greater haemodynamic stability whilst under the care of a vascular anaesthetist may reduce myocardial stress and subsequently asymptomatic damage, thereby improving medium-term prognosis. The database contained the maximum recorded intra-operative heart rate and minimum recorded systolic blood pressure for each patient. Surprisingly, elective patients cared for by a vascular anaesthetist tended to have a lower mean systolic blood pressure (120 versus 135mmHg; P<0.0001) and a higher maximum heart rate (77 versus 73 beats per minute; P=0.0002). However, elective vascular anaesthesia patients had a lower range of systolic blood pressures (145 versus 290mmHg) although the ranges for heart rate were similar (88 versus 90 beats per minute). The lower blood pressure range in the vascular anaesthesia patients may reflect tighter blood pressure control, despite the lower mean systolic.
Could the apparent benefit from vascular anaesthesia simply be a surrogate marker of simple differences in other aspects of care? The database does not contain pre-operative medications, which may have confounded the results. Peri-operative beta-blockade is not routinely used in Norwich. Patients established on beta-blockers continue them peri-operatively, but not routinely commenced pre-operatively. Statins are routinely started in all major vascular patients peri-operatively, if they are not already taking them. Epidural anaesthesia may improve outcome, but there was no difference in the frequency of use of epidural anaesthesia between vascular and non-vascular anaesthetists (47% versus 42%; P=0.16).
If repeated in other series, our findings have implications for service provision, suggesting that all elective major vascular surgery patients should be cared for anaesthetists with a special interest. If larger datasets identify a similar pattern in emergency vascular surgery, then separate vascular anaesthetic on-call rotas would be required. In the meantime, it should be noted that vascular surgical emergencies often require prompt surgical intervention to save life or limb. In such circumstances, intervention should not be delayed whilst seeking specialist anaesthetic support.
References
- . Effect of surgeon training, specialization, and experience on outcomes for cancer surgery: a systematic review of the literature. Ann Surg Oncol. 2009;16(7):1799–1808
- . Evolution of the surgeon-volume, patient-outcome relationship. Ann Surg. 2009;250(1):159–165
- . Meta-analysis and systematic review of the relationship between hospital volume and outcome following carotid endarterectomy. Eur J Vasc Endovasc Surg. 2007;33(6):645–651
- . Meta-analysis and systematic review of the relationship between volume and outcome in abdominal aortic aneurysm surgery. Br J Surg. 2007;94(4):395–403
- . Provider volume and outcomes for abdominal aortic aneurysm repair, carotid endarterectomy, and lower extremity revascularization procedures. J Vasc Surg. 2007;45(3):615–626
- Trends, complications, and mortality in peripheral vascular surgery. J Vasc Surg. 2006;43(2):205–216
- . Long-term survival after vascular surgery: specific influence of cardiac factors and implications for preoperative evaluation. J Vasc Surg. 2004;40(4):752–760
- . Perioperative- and long-term mortality rates after major vascular surgery: the relationship to preoperative testing in the medicare population. Anesth Analg. 1999;89(4):849–855
- . Abdominal aortic aneurysm: a service in need of surgery?. London: NCEPOD; 2005;www.ncepod.org.uk
- . Long-term consequences of anesthetic management. Anesthesiology. 2009;111(1):1–4
- . Effects of increased vascular surgical specialization on general surgery trainees, practicing surgeons, and the provision of vascular surgical care. J Am Coll Surg. 2009;208(5):692–697
- . Perioperative asymptomatic cardiac damage after endovascular abdominal aneurysm repair is associated with poor long-term outcome. J Vasc Surg. 2009;50(4):749–754
- . Long-term prognostic value of asymptomatic cardiac troponin T elevations in patients after major vascular surgery. J Vasc Endovasc Surg. 2004;28(1):59–66
PII: S1078-5884(10)00110-3
doi:10.1016/j.ejvs.2010.02.004
© 2010 European Society for Vascular Surgery. Published by Elsevier Inc. All rights reserved.
Volume 39, Issue 6 , Pages 719-725, June 2010
