Advertisement

Editor's Choice – Volume–Outcome Relationships in Elective Abdominal Aortic Aneurysm Surgery: Analysis of the UK Hospital Episodes Statistics Database for the Getting It Right First Time (GIRFT) Programme

Open ArchivePublished:August 14, 2020DOI:https://doi.org/10.1016/j.ejvs.2020.07.069

      Objective

      To investigate whether a volume–outcome relationship exists for elective abdominal aortic aneurysm (AAA) surgery conducted within the National Health Service (NHS) in England.

      Methods

      This was an analysis of administrative data. Data were extracted from the Hospital Episodes Statistics database for England from April 2011 to March 2019 for all adult admissions for elective infrarenal AAA surgery. Data were extracted for the NHS trust and surgeon undertaking the procedure, the surgical technique used (open or endovascular), the financial year of admission, length of hospital and critical care stay during the procedure and subsequent emergency re-admissions (primary outcome) and deaths within 30 days. Multilevel modelling was used to adjust for hierarchy and confounding.

      Results

      A dataset of 31 829 procedures (8867 open, 22 962 endovascular) was extracted. For open surgery, lower trust annual volume was associated with higher 30 day emergency re-admission rates and higher 30 day mortality. For open surgery, lower surgeon annual volume was associated with higher 30 day mortality and length of hospital stay greater than the median. For endovascular surgery, lower surgeon annual volume was associated with not having an overnight stay in critical care. None of the other volume–outcome relationships investigated was significant.

      Conclusion

      For elective infrarenal AAA surgery in the UK NHS, there was strong evidence of a volume–outcome relationship for open surgery. However, evidence for a volume–outcome relationship is dependent on the specific procedure undertaken and the outcome of interest.

      Keywords

      Volume–outcome relationships in abdominal aortic aneurysm surgery in England were investigated in the light of the introduction of a screening programme for older males, recent initiatives to centralise services, and greater use of endovascular techniques. Following these changes, good evidence was found of a volume–outcome relationship for open surgery but not for endovascular aneurysm repair.

      Introduction

      The National Health Service (NHS) in England strives to offer the best possible care within a finite budget.
      • NHS England
      NHS Long Term Plan.
      As pressure on services grows, continual review of quality and efficiency is essential. One area of particular interest has been a desire to centralise certain services to larger providers, where there is evidence that outcomes may be better for patients when a particular procedure is performed in high volume.
      • Imison C.
      • Sonola L.
      • Honeyman M.
      • Ross S.
      The Reconfiguration of Clinical Services. What Is the Evidence?.
      In England, centralisation of vascular surgical services into a hub and spoke model (with smaller “spoke” hospitals feeding patients into a larger “hub” for certain types of surgery) was originally driven by the need to have a comprehensive round the clock service for urgent and emergency vascular procedures. This approach was supported by the hypothesis that vascular surgical departments which perform large numbers of a particular procedure may be better at that procedure than trusts or surgeons who perform the procedure less frequently.
      • Halm E.A.
      • Lee C.
      • Chassin M.R.
      How is volume related to outcome in health care? A systematic review of the research literature.
      However, a clear volume–outcome relationship has only been established for certain types of procedures and, even where a relationship has been established, it appears dependant on the outcome investigated and the criteria used to define high and low volume.
      • Halm E.A.
      • Lee C.
      • Chassin M.R.
      How is volume related to outcome in health care? A systematic review of the research literature.
      There is evidence that the procedure in question must be of relatively high complexity for a relationship to exist.
      • Nimptsch U.
      • Mansky T.
      Hospital volume and mortality for 25 types of inpatient treatment in German hospitals: observational study using complete national data from 2009 to 2014.
      ,
      • Morche J.
      • Mathes T.
      • Pieper D.
      Relationship between surgeon volume and outcomes: a systematic review of systematic reviews.
      Furthermore, there are concerns that over centralisation could result in a more fragile healthcare system, which is less flexible, less able to cope with stressors, and where training opportunities are limited to specialist centres.
      • Imison C.
      • Sonola L.
      • Honeyman M.
      • Ross S.
      The Reconfiguration of Clinical Services. What Is the Evidence?.
      The Getting It Right First Time (GIRFT) programme is funded by the Department of Health in England with a remit to reduce unwarranted variation in patient outcomes within the NHS. Within the field of vascular surgery one source of such variation is thought to be differences in practice between high and low volume trusts and surgeons.
      • Horrocks M.
      Vascular Surgery: GIRFT programme national specialty report.
      Building on recommendations by The Vascular Society of Great Britain and Ireland,
      The Vascular Society of Great Britain and Ireland
      The Provision of Services for Patients with Vascular Disease.
      the GIRFT vascular surgery national report from 2018 made a specific recommendation around minimum annual surgical volumes for abdominal aortic aneurysm (AAA) of 60 procedures per year per trust, and that each trust must be staffed by a minimum of six vascular surgeons and six vascular interventional clinicians; either interventional radiologists or appropriately trained vascular surgeons.
      • Horrocks M.
      Vascular Surgery: GIRFT programme national specialty report.
      There is substantial previous research to suggest that there is a relationship between annual volume of AAA surgery and mortality.
      • Holt P.J.
      • Poloniecki J.D.
      • Gerrard D.
      • Loftus I.M.
      • Thompson M.M.
      Meta-analysis and systematic review of the relationship between volume and outcome in abdominal aortic aneurysm surgery.
      • Holt P.J.
      • Poloniecki J.D.
      • Loftus I.M.
      • Michaels J.A.
      • Thompson M.M.
      Epidemiological study of the relationship between volume and outcome after abdominal aortic aneurysm surgery in the UK from 2000 to 2005.
      • Landon B.E.
      • O'Malley A.J.
      • Giles K.
      • Cotterill P.
      • Schermerhorn M.L.
      Volume–outcome relationships and abdominal aortic aneurysm repair.
      • Trenner M.
      • Kuehnl A.
      • Salvermoser M.
      • Reutersberg B.
      • Geisbuesch S.
      • Schmid V.
      • et al.
      Editor's choice - high annual hospital volume is associated with decreased in hospital mortality and complication rates following treatment of abdominal aortic aneurysms: secondary data analysis of the nationwide German DRG statistics from 2005 to 2013.
      • Sawang M.
      • Paravastu S.C.V.
      • Liu Z.
      • Thomas S.D.
      • Beiles C.B.
      • Mwipatayi B.P.
      • et al.
      The relationship between aortic aneurysm surgery volume and peri-operative mortality in Australia.
      • Scali S.T.
      • Beck A.W.
      • Sedrakyan A.
      • Mao J.
      • Venermo M.
      • Faizer R.
      • et al.
      Hospital volume Association with abdominal aortic aneurysm repair mortality: analysis of the international Consortium of vascular Registries.
      However, in England, an AAA screening programme was introduced in 2009 for men aged ≥ 65 years, and UK outcome data related to surgical volume from after this period are lacking.
      • Holt P.J.
      • Poloniecki J.D.
      • Loftus I.M.
      • Michaels J.A.
      • Thompson M.M.
      Epidemiological study of the relationship between volume and outcome after abdominal aortic aneurysm surgery in the UK from 2000 to 2005.
      There is strong evidence that AAA screening programmes have had a positive effect on overall mortality from AAAs, with fewer deaths following rupture.
      • Guirguis-Blake J.M.
      • Beil T.L.
      • Senger C.A.
      • Whitlock E.P.
      Ultrasonography screening for abdominal aortic aneurysms: a systematic evidence review for the U.S. Preventive Services Task Force.
      ,
      • Jacomelli J.
      • Summers L.
      • Stevenson A.
      • Lees T.
      • Earnshaw J.J.
      Impact of the first 5 years of a national abdominal aortic aneurysm screening programme.
      Screening can also impact on peri-operative and short term mortality after elective surgery, with aneurysms being repaired earlier, and reduced subsequent emergency admissions.
      • Guirguis-Blake J.M.
      • Beil T.L.
      • Senger C.A.
      • Whitlock E.P.
      Ultrasonography screening for abdominal aortic aneurysms: a systematic evidence review for the U.S. Preventive Services Task Force.
      ,
      • Wanhainen A.
      • Hultgren R.
      • Linne A.
      • Holst J.
      • Gottsater A.
      • Langenskiold M.
      • et al.
      Outcome of the Swedish nationwide abdominal aortic aneurysm screening program.
      The overall fall in mortality following AAA surgery in recent years has also been driven by the increasing use of endovascular aneurysm repair (EVAR), which has a much lower mortality rate than open surgery.
      • Jacomelli J.
      • Summers L.
      • Stevenson A.
      • Lees T.
      • Earnshaw J.J.
      Update on the prevention of death from ruptured abdominal aortic aneurysm.
      It is not clear how these changes have impacted the association between volume and outcomes for elective AAA surgery.
      The aim of this study was to investigate the nature and extent of any volume–outcome relationship in elective AAA surgery conducted in England following these changes in practice, and to test the specific GIRFT recommendation of 60 procedures per trust per year and the implied recommendation of 10 procedures per surgeon per year.

      Methods

      Ethics

      This study used data from the UK Hospital Episode Statistics (HES) and Office for National Statistics (ONS) databases. The presentation of data follows current NHS Digital guidance for use of such data for research purposes.
      • NHS Digital
      Hospital Episode statistics (HES) analysis guide.
      Consent from individuals involved in this study was not required because it was an analysis of routine clinical data. Ethical approval was not sought for the present study because it did not directly involve human participants. This study was completed in accordance with the Helsinki Declaration as revised in 2013.

      Setting, timing, and participants

      The HES database contains data collected for patients seen at hospitals in England where the care is funded by the NHS. All NHS hospitals in England are run by trusts. Each trust covers a geographically defined catchment area of varying physical size and population. A single trust typically runs between one and four large secondary or tertiary care hospitals.
      The data presented here are taken from the HES database for the eight year period from 1 April 2011 to 31 March 2019, with 30 day follow up for subsequent emergency re-admissions and deaths to the end of April 2019 and 90 day follow up for subsequent admissions for a further AAA procedure to the end of June 2019. This time period was used to ensure that the data were recent enough to be relevant to current clinical practice.
      • Halm E.A.
      • Lee C.
      • Chassin M.R.
      Is volume related to outcome in health care? A systematic review and methodologic critique of the literature.
      The HES data were linked to UK ONS data on all cause mortality via a pseudonymised patient identifier.

      Inclusion and exclusion criteria

      Inclusion

      Infrarenal AAA operations (open or EVAR techniques) were extracted using the Classification of Interventions and Procedures codes (OPCS) version 4 codes: L184, L185, L186, L188, L189, L194, L195, L196, L198, L199, L265, L266, L267, L271, L275, L276, L278, L279, L281, L284, L285, L286, L288, L289.

      Exclusions

      Age <18 years; ruptured AAAs (ICD-10 codes I711, I713, I715, I718); emergency admissions.
      The procedure descriptions for these OPCS and ICD codes are given in Table S1. To ensure all datapoints were independent of one another at a patient level, only the chronologically first recorded procedure for anyone who had more than one such AAA procedure was included in the dataset.

      Data extracted

      Primary outcomes

      30 day post-discharge emergency hospital re-admissions. This was chosen as the primary outcome, as an emergency re-admission within 30 days of the index procedure is likely to be related to a serious complication of that procedure. Thus, it is a useful proxy measure for a serious complication. Given falling mortality rates following elective AAA surgery, particularly for EVAR, this was felt to be a more meaningful primary outcome measure with regard to population impact.

      Secondary outcomes

      All cause mortality during the index stay or at 30 days from discharge; length of hospital stay; use of critical care facilities during stay; and re-admission for a further infrarenal AAA procedure within 90 days of discharge. Further infrarenal AAA procedures within 90 days were identified using the same OPCS codes as for the index procedure but included ruptures and emergency and day -case admissions. They were linked at a patient level using the pseudonymised patient identifier. Ninety days was preferred to 30 days for this outcome to allow for possible delays to re-intervention where it was deemed non-urgent. An additional long term analysis of subsequent infrarenal AAA procedures at two years was conducted on the subset of the study cohort who had an index procedure before 30 June 2017.

      Primary exposure

      Annual volume of open or EVAR procedures conducted for each trust and for each surgeon. Annual volumes were calculated by taking the total number of elective infrarenal AAA procedures (regardless of whether an EVAR or open technique was used) conducted by a surgeon or trust over the study period and dividing it by the number of financial years that each surgeon or trust contributed data. It was chosen to define volume in terms of the volume of both procedures, in line with the GIRFT, Vascular Society of Great Britain and Ireland and European Society for Vascular Surgery volume definition.
      Annual volume data were then dichotomised to test the primary hypothesis (<10 procedures per annum for surgeon volume and <60 procedures per annum for trust volume). These cut offs are taken from the GIRFT national report and the Vascular Society report from 2018.
      • Horrocks M.
      Vascular Surgery: GIRFT programme national specialty report.
      ,
      The Vascular Society of Great Britain and Ireland
      The Provision of Services for Patients with Vascular Disease.
      In secondary analyses, surgeon annual volume was categorised as < 5, 5–9, 10–14, 15–19, and ≥20, and trust volumes were categorised as < 20, 20–39, 40–59, 60–79, 80–99, and ≥100 based on visual inspection of the data. Categorisation of annual volume was preferred over analysis of volume of a continuous variable after visual inspection of the relationship between volume and the outcomes investigated suggested that in many cases any relationship was likely to be inconsistent, complicating interpretation of the findings.
      Covariates were patient age (five year age bands), sex, Hospital Frailty Risk Score (HFRS) band (none, mild, moderate, severe),
      • Gilbert T.
      • Neuburger J.
      • Kraindler J.
      • Keeble E.
      • Smith P.
      • Ariti C.
      • et al.
      Development and validation of a Hospital Frailty Risk Score focusing on older people in acute care settings using electronic hospital records: an observational study.
      and year of admission. Age was categorised into five year age bands based on visual inspection of the data, to create broadly equal categories. As for volume, categorisation was preferred over treating age as a continuous variable because of concerns over the consistency of the relationship. HFRS score is calculated from ICD-10 data entered into HES, with 109 ICD-10 codes used to identify levels of frailty.
      • Gilbert T.
      • Neuburger J.
      • Kraindler J.
      • Keeble E.
      • Smith P.
      • Ariti C.
      • et al.
      Development and validation of a Hospital Frailty Risk Score focusing on older people in acute care settings using electronic hospital records: an observational study.

      Data management and statistical analyses

      Data were extracted from and onto a secure, encrypted server. Data were analysed within this secure environment using standard statistical software: Microsoft Excel (Microsoft Corp, Redmond, WA, USA), SAS (SAS Institute Inc, Cary, NC, USA), and Alteryx (Alteryx Inc, Irvine, CA, USA). Standard descriptive statistics (e.g. mean, median, frequency) are used as appropriate to the level of the data.
      Multilevel modelling was used to adjust for the hierarchical nature of the data using the GLIMMIX protocol in SAS. Two level (patients nested within trusts) intercept only models, with a logit link function were constructed for each outcome. For the outcome length of stay, the median length of stay was used to dichotomise the data for each procedure (more than eight days for open, more than three days for EVAR). For the outcome use of critical care facilities, those with any overnight stay in critical care and those without an overnight stay in critical care were compared. Each model was adjusted for the potential confounders listed above. Year of surgery was included as a covariable to adjust for temporal trends. The results of the analysis are presented in terms of odds ratios (ORs), 95% confidence intervals (CIs), and corresponding tests of significance.

      Results

      During the study period 44 750 infrarenal AAA procedures were conducted in patients aged ≥18 years. Exclusion of 12 162 procedures as non-elective (coded as emergency admission and/or rupture) and 759 as subsequent procedures in the same patient, gave an analysis dataset of 31 829 (8 867 open, 22 962 EVAR) across 101 trusts (95 performing open, 93 performing EVAR).
      Seventy-eight trusts conducted ≥40 procedures during the study period and the variation in the proportion of procedures conducted as EVAR in these trusts is summarised in Fig. 1. Within these 78 trusts, the highest proportion of procedures conducted as EVAR was 100% and 19 (24.4%) trusts conducted > 80% of procedures as EVAR. The lowest proportion of procedures conducted as EVAR was 5.0% and seven (9.0%) trusts conducted < 50% of their procedures as EVAR.
      Figure 1
      Figure 1Variation in percentage of total 31 829 procedures conducted as endovascular abdominal aortic aneurysm (AAA) repair across the 78 National Health Service (NHS) trusts in England that conducted more than 40 procedures during the study period of from April 2011 to March 2019.
      For the full dataset, the variation in use of EVAR and open procedures, together with outcome data for emergency re-admission and mortality per financial year is summarised in Table 1. The most obvious trend is for increasing centralisation of both open and EVAR surgery within larger volume trusts, with a large reduction in the number of trusts performing <20 procedures per annum over the study period. Mortality rates also appear to be falling with time for both EVAR and open surgery.
      Table 1Variation in trust volumes and outcomes over time for open and endovascular abdominal aortic aneurysm repair in England from April 2011 to March 2019
      Procedure typeFinancial yearPatientsTrustsMean volume per trustTrusts performing ≥60 procedures
      Volumes are based on those for the specified year and procedure. For all other analyses, volumes are based on averages across all years and for the combined volume of open and EVAR surgery.
      Trusts performing <20 procedures
      Volumes are based on those for the specified year and procedure. For all other analyses, volumes are based on averages across all years and for the combined volume of open and EVAR surgery.
      30 day emergency re-admission30 day mortality
      Endovascular2011–123 0308834.41427301 (9.9)44 (1.5)
      2012–132 8878235.21124329 (11.4)28 (1.0)
      2013–143 0757938.91525342 (11.1)53 (1.7)
      2014–152 9987440.51516316 (10.5)43 (1.4)
      2015–163 0567242.41714322 (10.5)28 (0.9)
      2016–172 9296743.71813293 (10.0)27 (0.9)
      2017–182 7136541.71710261 (9.6)31 (1.1)
      2018–192 2746833.4919228 (10.0)20 (0.9)
      Open2011–121 4199514.9169101 (7.1)77 (5.4)
      2012–131 3118814.9062123 (9.4)62 (4.7)
      2013–141 2597616.6053106 (8.4)53 (4.2)
      2014–151 1427415.405387 (7.6)40 (3.5)
      2015–169206713.705970 (7.6)42 (4.6)
      2016–179456215.204584 (8.9)24 (2.5)
      2017–189365915.904178 (8.3)38 (4.1)
      2018–199356015.604278 (8.3)37 (4.0)
      Data are presented as n (%) unless stated otherwise. CI = confidence interval.
      Volumes are based on those for the specified year and procedure. For all other analyses, volumes are based on averages across all years and for the combined volume of open and EVAR surgery.
      The characteristics of the patients within the dataset are presented in Table 2. Those who had EVAR were noticeably older than those undergoing open surgery, although this was not reflected in frailty levels, which were similar for both procedures. Thirty day emergency re-admission rates and 90 day and two year re-operation rates were noticeably higher for EVAR. Mean length of hospital and critical care stay were less than half for EVAR compared with open surgery. Despite the greater age of those undergoing EVAR, 30 day mortality rates were more than three times higher for patients undergoing open surgery.
      Table 2Summary data for 8 867 open and 22 962 endovascular abdominal aortic aneurysm (AAA) repairs in England from April 2011 to March 2019
      Procedure typeEndovascular

      (n = 22 962)
      Open

      (n = 8 867)
      Number of trusts9593
      Number in each age band – y
       <651 590 (6.9)1 477 (16.7)
       65–693 280 (14.3)2 453 (27.7)
       70–744 731 (20.6)2 149 (24.2)
       75–796 069 (26.4)1 758 (19.8)
       ≥807 292 (31.8)1 030 (11.6)
      Number of males20 197 (87.9)7 878 (88.8)
      Number in each Hospital Frailty Risk Score category
       None9 497 (41.4)3 736 (42.1)
       Mild8 704 (37.9)3 109 (35.1)
       Moderate4 275 (18.6)1 904 (21.5)
       Severe486 (2.1)118 (1.3)
      Number in each trust annual volume category
       <20522 (2.3)487 (5.5)
       20–393 741 (16.3)1 568 (17.7)
       40–594 160 (18.1)1 714 (19.3)
       60–797 501 (32.7)2 712 (30.6)
       80–992 362 (10.3)1 334 (15.0)
       ≥1004 676 (20.4)1 052 (11.9)
      Number in each surgeon annual volume category
       <51 657 (7.2)995 (11.2)
       5–96 578 (28.6)2 994 (33.8)
       10–147 177 (31.3)2 721 (30.7)
       15–193 416 (14.9)1 007 (11.4)
       ≥204 069 (17.7)1 137 (12.8)
      No consultant recorded65 (0.3)13 (0.1)
      30 day emergency hospital re-admission2 392 (10.4)727 (8.2)
      30 day mortality274 (1.2)373 (4.2)
      Mean length of hospital stay ± SD – days4.5 ± 6.310.7 ± 11.4
      Mean length of critical care stay ± SD – days0.5 ± 1.62.0 ± 4.1
      90 day re-admission for further infrarenal AAA surgery151 (0.7)8 (0.1)
      Two year re-admission for further infrarenal AAA surgery408/18 656 (2.2)22/7 220 (0.3)
      Data are presented as n (%) unless stated otherwise. SD = standard deviation.
      Fig. 2 summarises the unadjusted relationship between the primary outcome measure (30 day emergency re-admission rate) and the five secondary outcome measures and trust and surgeon volume. Multilevel modelling was used to investigate these associations further by adjusting for the potentially confounding influence of age, sex, frailty, and time period of admission on outcomes. Odds ratios for the independent association of trust and surgeon volume with outcomes are presented in Table 3. For open surgery, lower trust annual volume was associated with higher 30 day emergency re-admission rates and higher 30 day mortality. For open surgery, lower surgeon annual volume was associated with higher 30 day mortality and length of hospital stay greater than the median. For EVAR, lower surgeon annual volume was associated with not having an overnight stay in critical care. None of the volume measures were associated with readmissions for further AAA surgery at either 90 days or two years.
      Figure 2
      Figure 2Variation in outcomes for endovascular and open abdominal aortic aneurysm (AAA) repair in England relative to surgeon and trust average annual volume categories (A) 30 day emergency hospital re-admission, (B) 30 day mortality, (C) length of hospital stay, (D) length of critical care stay, (E) 90 day return to hospital for a subsequent abdominal aortic aneurysm repair procedure, and (F) two year return to hospital for a subsequent abdominal aortic aneurysm repair procedure during the study period of from April 2011 to March 2019.
      Table 3Odds ratios for primary volume measures when added to multivariable multilevel models for the volume–outcome relationship in 22 062 endovascular and 8 867 open abdominal aortic aneurysm (AAA) repairs in England from April 2011 to March 2019
      Primary volume measureEndovascular AAA repairOpen AAA repair
      Odds ratio (95% CI)p valueOdds ratio (95% CI)p value
      30 day emergency hospital readmission
       <60 per annum per trust0.937 (0.837–1.050).261.290 (1.106–1.505).001
       <10 per annum per surgeon0.958 (0.873–1.052).371.094 (0.935–1.280).26
      30 day mortality
       <60 per annum per trust1.054 (0.819–1.356).691.419 (1.076–1.871).013
       <10 per annum per surgeon0.851 (0.657–1.101).221.273 (1.013–1.600).038
      Length of hospital stay greater than median
       <60 per annum per trust1.099 (0.738–1.637).641.083 (0.857–1.369).50
       <10 per annum per surgeon0.990 (0.923–1.062).791.122 (1.008–1.248).035
      Overnight critical care stay
       <60 per annum per trust0.971 (0.430–2.195).940.542 (0.272–1.083).083
       <10 per annum per surgeon0.805 (0.728–0.891)<.0011.234 (0.969–1.571).088
      90 day re-admission for infrarenal AAA surgery
       <60 per annum per trust1.213 (0.806–1.825).352.074 (0.414–10.426).37
       <10 per annum per surgeon
      p values < .050 indicate statistical significance at the 5% level. AAA = abdominal aortic aneurysm; CI = confidence interval.
      0.956 (0.674–1.355).802.068 (0.493–8.676).32
      Two year re-admission for infrarenal AAA surgery
       <60 per annum per trust0.879 (0.638–1.210).430.449 (0.157–12.830).14
       <10 per annum per surgeon
      p values < .050 indicate statistical significance at the 5% level. AAA = abdominal aortic aneurysm; CI = confidence interval.
      0.901 (0.717–1.133).371.892 (0.767–4.669).17
      p values < .050 indicate statistical significance at the 5% level. AAA = abdominal aortic aneurysm; CI = confidence interval.

      Discussion

      With the introduction of the AAA screening programme for men aged ≥65 years living in England, overall mortality from AAA rupture has fallen substantially and the proportion of people undergoing early elective AAA surgery has increased.
      • Jacomelli J.
      • Summers L.
      • Stevenson A.
      • Lees T.
      • Earnshaw J.J.
      Update on the prevention of death from ruptured abdominal aortic aneurysm.
      At the same time, use of EVAR surgery has become much more common. The present study investigated the nature of the volume–outcome relationship following these changes in service organisation and clinical practice. A significant relationship between patient outcomes and volume was evident for open surgery across a number of outcomes. For EVAR, the only significant association identified was between lower surgeon volumes and reduced overnight stay in critical care, partly reflecting service organisation rather than patient outcomes.
      Recent practice guidelines have discussed the association between the volume of elective AAA surgery conducted by a hospital or surgeon and outcomes for patients, and recommended minimum annual surgical volumes.
      • Horrocks M.
      Vascular Surgery: GIRFT programme national specialty report.
      ,
      The Vascular Society of Great Britain and Ireland
      The Provision of Services for Patients with Vascular Disease.
      ,
      • Wanhainen A.
      • Verzini F.
      • Van Herzeele I.
      • Allaire E.
      • Bown M.
      • Cohnert T.
      • et al.
      Editor's choice - European society for vascular surgery (ESVS) 2019 clinical practice guidelines on the management of abdominal aorto-iliac artery aneurysms.
      ,
      • Chaikof E.L.
      • Dalman R.L.
      • Eskandari M.K.
      • Jackson B.M.
      • Lee W.A.
      • Mansour M.A.
      • et al.
      The Society for Vascular Surgery practice guidelines on the care of patients with an abdominal aortic aneurysm.
      The European Society for Vascular Surgery suggest a minimum annual volume of 30 cases per year per centre, although they acknowledge that differences in service delivery between and within countries makes identifying a minimum volume difficult.
      • Wanhainen A.
      • Verzini F.
      • Van Herzeele I.
      • Allaire E.
      • Bown M.
      • Cohnert T.
      • et al.
      Editor's choice - European society for vascular surgery (ESVS) 2019 clinical practice guidelines on the management of abdominal aorto-iliac artery aneurysms.
      The US based Society for Vascular Surgery recommend a minimum annual volume per centre of 10 for EVAR and 10 for open surgery.
      • Chaikof E.L.
      • Dalman R.L.
      • Eskandari M.K.
      • Jackson B.M.
      • Lee W.A.
      • Mansour M.A.
      • et al.
      The Society for Vascular Surgery practice guidelines on the care of patients with an abdominal aortic aneurysm.
      Interestingly, these guidelines and the Vascular Society of Great Britain and Ireland
      The Vascular Society of Great Britain and Ireland
      The Provision of Services for Patients with Vascular Disease.
      guidelines recognise that, although there was good evidence of a volume–outcome relationship for elective AAA repair, the evidence for any particular cut off is weak, particularly as much of the cited research was conducted when EVAR use was much less common. As an example, a UK study used HES data from 2005–7 to investigate volume–outcome relationships for AAA repair and reported a significant relationship between hospital volume and in hospital mortality for both open and EVAR procedures.
      • Holt P.J.
      • Poloniecki J.D.
      • Khalid U.
      • Hinchliffe R.J.
      • Loftus I.M.
      • Thompson M.M.
      Effect of endovascular aneurysm repair on the volume–outcome relationship in aneurysm repair.
      However, of 7 313 elective AAA procedures included in the analysis, only 1 645 (22.5%) were EVAR, and reported mortality rates were noticeably higher than recorded in the present study, particularly for EVAR.
      That four of the five significant relationships identified in this study were for open surgery is perhaps unsurprising. Open AAA surgery is generally a more complex procedure than EVAR and its outcomes may be more dependent on the skill of the surgeon, the wider surgical team and the organisational structures within the trust.
      • Nimptsch U.
      • Mansky T.
      Hospital volume and mortality for 25 types of inpatient treatment in German hospitals: observational study using complete national data from 2009 to 2014.
      ,
      • Morche J.
      • Mathes T.
      • Pieper D.
      Relationship between surgeon volume and outcomes: a systematic review of systematic reviews.
      The present findings, with regard to the strong relationship between short term mortality and annual volume for open surgery, support much of the previous research in this area, which has focused on mortality as an outcome.
      • Nimptsch U.
      • Mansky T.
      Hospital volume and mortality for 25 types of inpatient treatment in German hospitals: observational study using complete national data from 2009 to 2014.
      ,
      • Holt P.J.
      • Poloniecki J.D.
      • Gerrard D.
      • Loftus I.M.
      • Thompson M.M.
      Meta-analysis and systematic review of the relationship between volume and outcome in abdominal aortic aneurysm surgery.
      ,
      • Trenner M.
      • Kuehnl A.
      • Salvermoser M.
      • Reutersberg B.
      • Geisbuesch S.
      • Schmid V.
      • et al.
      Editor's choice - high annual hospital volume is associated with decreased in hospital mortality and complication rates following treatment of abdominal aortic aneurysms: secondary data analysis of the nationwide German DRG statistics from 2005 to 2013.
      ,
      • Eckstein H.H.
      • Bruckner T.
      • Heider P.
      • Wolf O.
      • Hanke M.
      • Niedermeier H.P.
      • et al.
      The relationship between volume and outcome following elective open repair of abdominal aortic aneurysms (AAA) in 131 German hospitals.
      • Zettervall S.L.
      • Schermerhorn M.L.
      • Soden P.A.
      • McCallum J.C.
      • Shean K.E.
      • Deery S.E.
      • et al.
      The effect of surgeon and hospital volume on mortality after open and endovascular repair of abdominal aortic aneurysms.
      • Budtz-Lilly J.
      • Venermo M.
      • Debus S.
      • Behrendt C.A.
      • Altreuther M.
      • Beiles B.
      • et al.
      Editor's choice - assessment of international outcomes of intact abdominal aortic aneurysm repair over 9 years.
      • Dubois L.
      • Allen B.
      • Bray-Jenkyn K.
      • Power A.H.
      • DeRose G.
      • Forbes T.L.
      • et al.
      Higher surgeon annual volume, but not years of experience, is associated with reduced rates of postoperative complications and reoperations after open abdominal aortic aneurysm repair.
      • Phillips P.
      • Poku E.
      • Essat M.
      • Woods H.B.
      • Goka E.A.
      • Kaltenthaler E.C.
      • et al.
      Procedure volume and the association with short-term mortality following abdominal aortic aneurysm repair in European populations: a systematic review.
      A study of data from 11 countries for 82 253 patients over nine years from the International Consortium of Vascular Registries noted a volume–outcome relationship for open surgery, but not EVAR.
      • Budtz-Lilly J.
      • Venermo M.
      • Debus S.
      • Behrendt C.A.
      • Altreuther M.
      • Beiles B.
      • et al.
      Editor's choice - assessment of international outcomes of intact abdominal aortic aneurysm repair over 9 years.
      Interestingly, they also noted declining annual volumes of open surgery with time, and suggested that technical competency for open surgery may only be maintained in the long term by increased centralisation. In a large scale study from Germany, of 41 678 patients receiving EVAR for AAA from 2009–14, in hospital mortality rates in the highest and lowest volume quintiles were 1.6% and 1.7%, respectively, with no significant relationship between volume and mortality.
      • Nimptsch U.
      • Mansky T.
      Hospital volume and mortality for 25 types of inpatient treatment in German hospitals: observational study using complete national data from 2009 to 2014.
      However, for 22 227 patients receiving open surgery, there was a significant relationship, with mortality rates in the highest and lowest quintiles of 4.7% and 7.8%, respectively. A further study from Germany using data from 2005–13 found evidence of a link between hospital volume and mortality, complication rates and use of blood products for non-ruptured AAA surgery.
      • Trenner M.
      • Kuehnl A.
      • Salvermoser M.
      • Reutersberg B.
      • Geisbuesch S.
      • Schmid V.
      • et al.
      Editor's choice - high annual hospital volume is associated with decreased in hospital mortality and complication rates following treatment of abdominal aortic aneurysms: secondary data analysis of the nationwide German DRG statistics from 2005 to 2013.
      The authors proposed a minimum hospital volume of 75 procedures per annum. Finally, a review of 16 European studies was published in 2017.
      • Phillips P.
      • Poku E.
      • Essat M.
      • Woods H.B.
      • Goka E.A.
      • Kaltenthaler E.C.
      • et al.
      Procedure volume and the association with short-term mortality following abdominal aortic aneurysm repair in European populations: a systematic review.
      The authors concluded that there was evidence of a relationship between hospital, but not surgeon, volume and mortality, and no strong evidence of a relationship between any volume measure and other outcomes. Although the review considered only studies published in the preceding 10 years, the first year of included data for any of the studies was 1994 and the last year was 2013, with all studies including data for the period 2000–10. As such, their conclusions are likely to reflect earlier clinical practice.
      EVAR was associated with higher 30 day emergency re-admission and 90 day and two year re-intervention rates than open surgery. However, the two groups are likely to vary with regard to clinical presentation and anatomical features of the aneurysm. Patient preference and resource availability may also play a role. For these reasons, a direct comparison of outcomes between EVAR and open repair using this dataset is not presented. A recently published study used HES data to investigate long term outcomes after repair of unruptured AAAs in 37 138 patients and found that rupture rates were higher for EVAR.
      • Johal A.S.
      • Loftus I.M.
      • Boyle J.R.
      • Heikkila K.
      • Waton S.
      • Cromwell D.A.
      Long-term survival after endovascular and open repair of unruptured abdominal aortic aneurysm.
      There was notable variation in the use of EVAR and open surgery between trusts in the present dataset. Differences in presentation, anatomical features, resource availability and patient preference may also play a role here. Furthermore, some larger trusts may act as regional centres and so see more complex cases. Nevertheless, some of this variation is likely to reflect the preference of surgical teams. It was decided to stratify the present analyses by surgical technique to avoid this confounding the findings. Future clinical guidelines should look to set specific minimum volume thresholds for each technique separately, particularly as volume thresholds appear to be more pertinent to open surgery. A study that looks to define a minimum volume threshold for open AAA repair in isolation is merited.

      Strengths

      The present data cover all elective, infrarenal AAA procedures carried out by the NHS in England during an eight year period up to March 2019. All but a small minority of AAA procedures performed in England are funded by the NHS. Studies that use data from countries, such as the United States, where healthcare provision is more fragmented and based on private health insurance are likely to be less comprehensive and may be biased in terms of the demographic and socio–economic profile of their cohort.
      • Beck A.W.
      • Sedrakyan A.
      • Mao J.
      • Venermo M.
      • Faizer R.
      • Debus S.
      • et al.
      Variations in abdominal aortic aneurysm care: a report from the international Consortium of vascular Registries.
      The present dataset was also relatively homogenous. Only non-ruptured and elective admissions were included, thus minimising the impact of confounding.
      • Halm E.A.
      • Lee C.
      • Chassin M.R.
      Is volume related to outcome in health care? A systematic review and methodologic critique of the literature.
      Consideration of why any observed volume–outcome relationship exists is also important. A volume–outcome relationship can arise where higher volumes lead to more skilled and more efficient practice; the “practice makes perfect” model. In settings where healthcare provision is based on health insurance schemes, selective referral can make it more likely that a volume–outcome relationship will be observed, with centres with better outcomes receiving more referrals and so having higher volumes.
      • Halm E.A.
      • Lee C.
      • Chassin M.R.
      Is volume related to outcome in health care? A systematic review and methodologic critique of the literature.
      In this case good outcomes lead to higher volumes. Only if the practice makes perfect model is operating does it make sense to reorganise services to eliminate low volume providers. In the UK, active patient choice over healthcare provider is relatively uncommon.
      • Dixon A.
      • Robertson R.
      • Appleby J.
      • Burge P.
      • Devlin N.
      • Magee H.
      Patient Choice: how Patients Choose and How Providers Respond.
      As such, selective referral is unlikely to drive any observed volume–outcome relationship in the present dataset and the relationships seen can reasonably be attributed to higher volumes leading to better outcomes.

      Limitations

      It was not possible to adjust for aspects of clinical presentation, such as the size and nature of the aneurysm and it is acknowledged that this lack of co-morbidity and anatomical detail is a limitation to the findings. It should be emphasised that the present findings relate to elective infrarenal AAA only and should not be extrapolated to other procedures.
      Specific peri- and post-surgical complications could not be investigated as outcomes, as such data are not well defined in the HES dataset. Emergency re-admissions were used as a proxy measure for post-surgical complications in a general sense. Such admissions are likely to include those unrelated to the index procedure. However, by restricting follow up to 30 days, a high proportion of emergency readmissions are likely to relate to the index procedure, and those that do not are unlikely to vary systematically by annual surgical volume and so confound the present findings. It is also recognised that all of the outcomes could be influenced, to an extent, by administrative factors, such as how post-surgery care is organised locally. Nevertheless, the need for emergency re-admission, re-intervention, and excessive hospital stay are of direct relevance to patient quality of life.
      Coding errors within the HES dataset have been identified previously, and this is particularly pertinent to this dataset with regard to attribution of a surgeon to a procedure. However, it is felt that such errors will not have resulted in substantial systematic bias or in any change to the overall conclusions. Finally, it is acknowledged that had a different measure of volume and a different volume cut off been used, the present conclusions may have been different. The choice of high/low volume cut offs was based on the recommendations of the GIRFT report and the Vascular Society of Great Britain and Ireland.

      Conclusions

      It is important for healthcare service providers to have evidence on which to base decisions regarding service organisation. For AAA surgery, at the current level of centralisation, with recent increases in the use of EVAR surgery, lower mortality rates than seen historically, and with a national screening programme in place, there is evidence of a volume–outcome relationship for open surgery at the thresholds investigated. Although this was dependent on the outcomes investigated and whether trust or surgeon volume was of interest, this is consistent with previous research. For EVAR surgery a volume-patient outcome relationship was not evident. On this basis, the case for setting of minimum volume thresholds for open elective AAA repair appears to be strong and the specific thresholds investigated seem appropriate. Future research should investigate how best to measure surgical volume; whether open and EVAR should be considered together or separately, whether emergency admissions and ruptures should be included in the volume measure and over what time period volume should be considered. The perceptions and impacts of centralisation of services on patients, clinicians and health service managers should also be investigated.

      Conflict of Interest

      None.

      Funding

      None.

      Contributorship

      This study was designed and organised by MH, JD, and WKG. Data cleaning, analysis, and writing of the first draft was by WKG, supported by JD and MH. All authors critically reviewed the manuscript and agreed to submission of the final draft.

      Information Governance

      This report does not contain patient identifiable data. The data in this report are anonymised. Request for any underlying data will not be granted as the data are calculated from data under licence/data sharing agreement from NHS Digital and/or other data provider where conditions of use (and further use) apply. Copyright © 2013, 2014, 2015, 2016, 2017 re-used with the permission of the Health & Social Care Information Centre. All rights reserved.

      Acknowledgements

      We acknowledge The UK Office for National Statistics (ONS) and NHS Digital for permission to use their data in this report. We also thank all staff within individual NHS trusts who collected and entered the data used in this study.

      Appendix A. Supplementary data

      The following is the Supplementary data to this article:

      References

        • NHS England
        NHS Long Term Plan.
        NHS England, London, UK2019
        • Imison C.
        • Sonola L.
        • Honeyman M.
        • Ross S.
        The Reconfiguration of Clinical Services. What Is the Evidence?.
        The King’s Fund, London, UK2014
        • Halm E.A.
        • Lee C.
        • Chassin M.R.
        How is volume related to outcome in health care? A systematic review of the research literature.
        in: Hewitt M. Interpreting the volume–outcome relationship in the context of healthcare quality: workshop summary. 2002/09/17. National Academic Press, Washington DC2000: 511-520
        • Nimptsch U.
        • Mansky T.
        Hospital volume and mortality for 25 types of inpatient treatment in German hospitals: observational study using complete national data from 2009 to 2014.
        BMJ Open. 2017; 7e016184
        • Morche J.
        • Mathes T.
        • Pieper D.
        Relationship between surgeon volume and outcomes: a systematic review of systematic reviews.
        Syst Rev. 2016; 5: 204
        • Horrocks M.
        Vascular Surgery: GIRFT programme national specialty report.
        GIRFT/NHS Improvement, London, UK2018
        • The Vascular Society of Great Britain and Ireland
        The Provision of Services for Patients with Vascular Disease.
        The Vascular Society of Great Britain and Ireland, London, UK2018
        • Holt P.J.
        • Poloniecki J.D.
        • Gerrard D.
        • Loftus I.M.
        • Thompson M.M.
        Meta-analysis and systematic review of the relationship between volume and outcome in abdominal aortic aneurysm surgery.
        Br J Surg. 2007; 94: 395-403
        • Holt P.J.
        • Poloniecki J.D.
        • Loftus I.M.
        • Michaels J.A.
        • Thompson M.M.
        Epidemiological study of the relationship between volume and outcome after abdominal aortic aneurysm surgery in the UK from 2000 to 2005.
        Br J Surg. 2007; 94: 441-448
        • Landon B.E.
        • O'Malley A.J.
        • Giles K.
        • Cotterill P.
        • Schermerhorn M.L.
        Volume–outcome relationships and abdominal aortic aneurysm repair.
        Circulation. 2010; 122: 1290-1297
        • Trenner M.
        • Kuehnl A.
        • Salvermoser M.
        • Reutersberg B.
        • Geisbuesch S.
        • Schmid V.
        • et al.
        Editor's choice - high annual hospital volume is associated with decreased in hospital mortality and complication rates following treatment of abdominal aortic aneurysms: secondary data analysis of the nationwide German DRG statistics from 2005 to 2013.
        Eur J Vasc Endovasc Surg. 2018; 55: 185-194
        • Sawang M.
        • Paravastu S.C.V.
        • Liu Z.
        • Thomas S.D.
        • Beiles C.B.
        • Mwipatayi B.P.
        • et al.
        The relationship between aortic aneurysm surgery volume and peri-operative mortality in Australia.
        Eur J Vasc Endovasc Surg. 2019; 57: 510-519
        • Scali S.T.
        • Beck A.W.
        • Sedrakyan A.
        • Mao J.
        • Venermo M.
        • Faizer R.
        • et al.
        Hospital volume Association with abdominal aortic aneurysm repair mortality: analysis of the international Consortium of vascular Registries.
        Circulation. 2019; 140: 1285-1287
        • Guirguis-Blake J.M.
        • Beil T.L.
        • Senger C.A.
        • Whitlock E.P.
        Ultrasonography screening for abdominal aortic aneurysms: a systematic evidence review for the U.S. Preventive Services Task Force.
        Ann Intern Med. 2014; 160: 321-329
        • Jacomelli J.
        • Summers L.
        • Stevenson A.
        • Lees T.
        • Earnshaw J.J.
        Impact of the first 5 years of a national abdominal aortic aneurysm screening programme.
        Br J Surg. 2016; 103: 1125-1131
        • Wanhainen A.
        • Hultgren R.
        • Linne A.
        • Holst J.
        • Gottsater A.
        • Langenskiold M.
        • et al.
        Outcome of the Swedish nationwide abdominal aortic aneurysm screening program.
        Circulation. 2016; 134: 1141-1148
        • Jacomelli J.
        • Summers L.
        • Stevenson A.
        • Lees T.
        • Earnshaw J.J.
        Update on the prevention of death from ruptured abdominal aortic aneurysm.
        J Med Screen. 2017; 24: 166-168
        • NHS Digital
        Hospital Episode statistics (HES) analysis guide.
        NHS Digital, London, UK2018
        • Halm E.A.
        • Lee C.
        • Chassin M.R.
        Is volume related to outcome in health care? A systematic review and methodologic critique of the literature.
        Ann Intern Med. 2002; 137: 511-520
        • Gilbert T.
        • Neuburger J.
        • Kraindler J.
        • Keeble E.
        • Smith P.
        • Ariti C.
        • et al.
        Development and validation of a Hospital Frailty Risk Score focusing on older people in acute care settings using electronic hospital records: an observational study.
        Lancet. 2018; 391: 1775-1782
        • Wanhainen A.
        • Verzini F.
        • Van Herzeele I.
        • Allaire E.
        • Bown M.
        • Cohnert T.
        • et al.
        Editor's choice - European society for vascular surgery (ESVS) 2019 clinical practice guidelines on the management of abdominal aorto-iliac artery aneurysms.
        Eur J Vasc Endovasc Surg. 2019; 57: 8-93
        • Chaikof E.L.
        • Dalman R.L.
        • Eskandari M.K.
        • Jackson B.M.
        • Lee W.A.
        • Mansour M.A.
        • et al.
        The Society for Vascular Surgery practice guidelines on the care of patients with an abdominal aortic aneurysm.
        J Vasc Surg. 2018; 67: 2-77 e2
        • Holt P.J.
        • Poloniecki J.D.
        • Khalid U.
        • Hinchliffe R.J.
        • Loftus I.M.
        • Thompson M.M.
        Effect of endovascular aneurysm repair on the volume–outcome relationship in aneurysm repair.
        Circ Cardiovasc Qual Outcomes. 2009; 2: 624-632
        • Eckstein H.H.
        • Bruckner T.
        • Heider P.
        • Wolf O.
        • Hanke M.
        • Niedermeier H.P.
        • et al.
        The relationship between volume and outcome following elective open repair of abdominal aortic aneurysms (AAA) in 131 German hospitals.
        Eur J Vasc Endovasc Surg. 2007; 34: 260-266
        • Zettervall S.L.
        • Schermerhorn M.L.
        • Soden P.A.
        • McCallum J.C.
        • Shean K.E.
        • Deery S.E.
        • et al.
        The effect of surgeon and hospital volume on mortality after open and endovascular repair of abdominal aortic aneurysms.
        J Vasc Surg. 2017; 65: 626-634
        • Budtz-Lilly J.
        • Venermo M.
        • Debus S.
        • Behrendt C.A.
        • Altreuther M.
        • Beiles B.
        • et al.
        Editor's choice - assessment of international outcomes of intact abdominal aortic aneurysm repair over 9 years.
        Eur J Vasc Endovasc Surg. 2017; 54: 13-20
        • Dubois L.
        • Allen B.
        • Bray-Jenkyn K.
        • Power A.H.
        • DeRose G.
        • Forbes T.L.
        • et al.
        Higher surgeon annual volume, but not years of experience, is associated with reduced rates of postoperative complications and reoperations after open abdominal aortic aneurysm repair.
        J Vasc Surg. 2018; 67: 1717-17126 e5
        • Phillips P.
        • Poku E.
        • Essat M.
        • Woods H.B.
        • Goka E.A.
        • Kaltenthaler E.C.
        • et al.
        Procedure volume and the association with short-term mortality following abdominal aortic aneurysm repair in European populations: a systematic review.
        Eur J Vasc Endovasc Surg. 2017; 53: 77-88
        • Johal A.S.
        • Loftus I.M.
        • Boyle J.R.
        • Heikkila K.
        • Waton S.
        • Cromwell D.A.
        Long-term survival after endovascular and open repair of unruptured abdominal aortic aneurysm.
        Br J Surg. 2019; 106: 1784-1793
        • Beck A.W.
        • Sedrakyan A.
        • Mao J.
        • Venermo M.
        • Faizer R.
        • Debus S.
        • et al.
        Variations in abdominal aortic aneurysm care: a report from the international Consortium of vascular Registries.
        Circulation. 2016; 134: 1948-1958
        • Dixon A.
        • Robertson R.
        • Appleby J.
        • Burge P.
        • Devlin N.
        • Magee H.
        Patient Choice: how Patients Choose and How Providers Respond.
        The King's Fund, London, UK2010

      Linked Article

      • Readmission and Re-intervention are Better Measures of EVAR Quality
        European Journal of Vascular and Endovascular SurgeryVol. 60Issue 4
        • Preview
          Gray et al. 1 report a large dataset of 31 892 elective abdominal aortic aneurysm (AAA) repairs over an 8-year period from the Hospital Episode Statistics (HES) database in England. They explore the relationship between case volume and outcomes for patients undergoing open surgical repair (OSR) or endovascular aneurysm repair (EVAR) for infrarenal AAA.
        • Full-Text
        • PDF
        Open Archive

      Comments

      Commenting Guidelines

      To submit a comment for a journal article, please use the space above and note the following:

      • We will review submitted comments as soon as possible, striving for within two business days.
      • This forum is intended for constructive dialogue. Comments that are commercial or promotional in nature, pertain to specific medical cases, are not relevant to the article for which they have been submitted, or are otherwise inappropriate will not be posted.
      • We require that commenters identify themselves with names and affiliations.
      • Comments must be in compliance with our Terms & Conditions.
      • Comments are not peer-reviewed.