If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The Relationship Between Hospital Procedure Volume and Outcomes After Endovascular or Open Surgical Revascularisation for Peripheral Arterial Disease: An Analysis of Health Insurance Claims Data
Corresponding author. Head of Research Group GermanVasc, Department of Vascular and Endovascular Surgery, Asklepios Clinic Wandsbek, Asklepios Medical School, Hamburg, Germany.
Affiliations
Research Group GermanVasc, University Medical Centre Hamburg-Eppendorf, Hamburg, GermanyBrandenburg Medical School Theodor Fontane, Neuruppin, Germany
There is a paucity of data on the relationship between hospital procedure volume and outcomes after inpatient treatment of symptomatic peripheral arterial disease (PAD). This study aimed to generate meaningful hypotheses to support the ongoing discussion.
Methods
Data derived from BARMER, Germany’s second largest insurance provider, were linked with nationwide hospital procedure volumes from mandatory hospital quality reports. All endovascular (EVR) and open surgical revascularisations (OSR) provided to patients (≥ 40 years) with symptomatic PAD between 1 January 2013 and 31 December 2018 were included. Hospital volume was defined as the number of procedures performed by a hospital in the previous calendar year (in quartiles). Freedom from re-intervention, amputation, and overall mortality rate within 12 months after discharge were analysed using multivariable Cox proportional hazards models. In hospital mortality was determined by generalised estimating equations logistic regression models.
Results
There were 88 187 revascularisations (72.4% EVR; EVR: 72.7 years and 45.2% females; OSR: 71.9 years and 41.9% females) registered by 668 hospitals. No statistically significant association was found between 12 month freedom from re-intervention and hospital volume (EVR: 4; quartile HR 1.05; 95% CI 0.94 – 1.16. OSR: 4; quartile HR 1.05; 95% CI 0.92 – 1.21). Patients with OSR had a decreased hazard of 12 month mortality in a high volume hospital compared with a low volume hospital (HR 0.85; 95% CI 0.73 – 0.98), but not with EVR (HR 1.03; 95% CI 0.91 – 1.16). Patients who were treated in hospitals with highest volumes showed decreased hazards of 12 month freedom from amputation when compared with low volume hospitals (EVR: HR 0.72; 95% CI 0.52 – 0.99. OSR: HR 0.61; 95% CI 0.44 – 0.85).
Conclusion
This large retrospective analysis of insurance claims suggests that higher procedure volume is associated with lower major amputation rates, although there is a need for standardisation of the definition of volume stratification. Future studies should address the impact of subsequent outpatient care and surveillance to further examine the complex interaction between treatment and outcomes.
German hospital quality reports on annual procedure volume and longitudinal health insurance claims data providing outcomes beyond discharge were linked for the first time to compile a large cohort of 64 871 individual patients who underwent 88 187 hospitalisations for symptomatic peripheral arterial disease in Germany. While previous studies have been mostly limited to peri-procedural short term outcomes, the current study additionally explored the complex relationship between hospital volume and midterm outcomes after one year, indicating that higher volume was associated with reduced risk of amputation after both endovascular and open surgical revascularisation and with a lower mortality risk after open surgical revascularisation.
Introduction
The impact of hospital procedure and case volumes on outcomes has been extensively discussed for several cardiovascular populations.
Editor's Choice - The impact of centralisation and endovascular aneurysm repair on treatment of ruptured abdominal aortic aneurysms based on international registries.
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.
while the available evidence concerning interventions for patients with lower extremity peripheral arterial disease (PAD) remains scarce. Moreover, most conclusions in previous publications were derived from administrative and clinical registries, which were often limited to peri-procedural short term outcome data. That might be more acceptable after certain procedures, but it seems important to compare midterm outcomes after invasive treatment of patients with symptomatic PAD.
Long term evaluation should be an integral part of the clinical implementation of new vascular treatments - an ESVS Executive Committee position statement.
Editor's Choice - International variations and sex disparities in the treatment of peripheral arterial occlusive disease: a report from VASCUNET and the International Consortium of Vascular Registries.
PAD plays an important part in the everyday clinical practice of vascular surgeons and other medical specialties involved in its treatment. Considering the huge disease burden for global healthcare systems, it seems interesting that only a few and mainly historical studies have addressed the volume–outcome relationship in the PAD field with inconsistent conclusions, while valid guidelines do not contain specific recommendations.
Impact of institutional volume on critical in-hospital complications adjusted for patient- and limb-related characteristics: an analysis of a nationwide Japanese registry of endovascular interventions for PAD.
Editor's Choice - 2017 ESC Guidelines on the Diagnosis and Treatment of Peripheral Arterial Diseases, in collaboration with the European Society for Vascular Surgery (ESVS).
Beyond the unequivocal meaning of patient reported outcomes for the treatment of this patient population, major adverse limb and cardiovascular events count as commonly accepted indicators of outcome quality after invasive revascularisation for symptomatic PAD.
Random sampling and additional needs based reviews by the medical service of German statutory health insurance providers (Medizinischer Dienst der Krankenversicherung, MDK) were regularly applied to assure high internal and external data validity.
Hospital procedure volume
Hospital procedure volumes were extracted from mandatory hospital quality reports covering all facilities in Germany. German hospitals were obliged to submit a biennial hospital specific quality report between 2005 – 2014; this report has been requested annually by the authorities since 2014. This information is being increasingly used for analysing the relationship between hospital volume and health outcomes.
Hospital volume and mortality for 25 types of inpatient treatment in German hospitals: observational study using complete national data from 2009 to 2014.
Amongst general information on the hospitals, such as staffing, number of beds, and total number of patients treated, the annual number of procedures performed by operation and procedure codes and international classification of disease diagnoses (ICD-GM codes) were extracted from these reports. The selected operation and procedure codes for both endovascular (EVR) and open surgical revascularisation (OSR) were summed, respectively, for each clinic and reporting year as the annual procedure volume. The annual procedure volume for each procedure type (EVR vs. OSR) was divided into quartiles by sorting the procedure volume in ascending order and separating the cumulative sum to generate four groups of equal size. Resulting quartile thresholds varied between years and are displayed in Supplementary Fig. S1.
Study population
Hospital cases were selected from the database of BARMER, Germany’s second largest insurance provider. The database comprises longitudinal inpatient and outpatient information of about 9 million insured persons and is widely used for health services research.
Editor's Choice - long term survival after femoropopliteal artery revascularisation with paclitaxel coated devices: a propensity score matched cohort analysis.
A retrospective cohort study on the provision and outcomes of pharmacological therapy after revascularization for peripheral arterial occlusive disease: A study protocol.
Inclusion criteria included any in hospital stay for symptomatic PAD treated by either an EVR or OSR (see Supplementary Table S1 for specific coding) between January 1 2013 and December 31 2018. Eligible PAD patients were categorised in Fontaine stages 2, 3, and 4. To establish a more homogeneous cohort, only patients aged ≥ 40 years on admission were included. The longitudinally linked data were collected for each patient on a hospital case level, enabling inclusion of all eligible cases (hospitalisations) per patient. To adjust for case severity, longitudinal information on medical history up to five years before each hospital stay was extracted from previous inpatient and outpatient claims data (lookback). Hybrid cases (both EVR and OSR during the same procedure), cases with incomplete five year lookback, and cases without specification of the revascularised leg or lesion level were excluded.
Data linkage
The nine digit institution code provides a unique identifier for each hospital in Germany and was used in this study to interlink information extracted from hospital quality reports covering all hospitals and claims data. For each case, the hospital information of the previous calendar year was used. Cases with missing information due to incomplete data on the previous year (e.g., newly founded hospital, mergers, or acquisitions with change in institution code) were excluded.
Declaration of Helsinki
The study complied with the Declaration of Helsinki 2013. All analyses were based on a factual anonymised administrative database so that neither locally appointed ethics committee approval nor informed consent was applicable.
Study outcomes
Study outcomes were 12 month freedom from re-intervention (defined as subsequent EVR or OSR in the index leg and same lesion level). Lesion levels were categorised in aorto-iliac, femoropopliteal, and crural (below the knee) revascularisations. Additional outcomes were 12 month freedom from major amputation, 12 month overall survival after discharge, and in hospital mortality. An amputation was considered if it affected the index leg or either leg if the patient had procedures on both legs during the hospital stay.
Statistical analysis
To detect erroneous outliers in hospital quality reports, the hospital procedure volume and total number of patients treated were studied for each hospital over time using modified Z scores with a critical value of 3.5.
Hospitals with Z scores above the critical value were excluded for the specific year (1.7% of hospitals).
For a summary of baseline characteristics, continuous variables are shown as mean ± standard deviation or median with interquartile range depending on their distribution. Categorical variables are presented as percentages. Cases with missing data in general patient information were excluded using listwise deletion, resulting in complete case analysis (< 0.5% of all cases).
The outcomes were analysed using separate multivariable Cox proportional hazards models clustered by hospital and stratified by Fontaine stage. The proportional hazards assumption was assessed graphically and tested with the Grambsch–Therneau test.
Analysis of freedom from major amputation included only patients without prior major amputation and was performed at patient level using only the first inpatient admission during the study time to avoid an overestimation of amputation events. Patients who had changed their insurance provider during follow up were considered right censored in the Cox proportional hazards models. In hospital death was examined using generalised estimating equations in the form of logistic models with binomial variance functions to account for observations clustered by hospital.
Each model was adjusted for the patient’s sex, number of prescribed medications in the year prior to admission, previous prescriptions for antithrombotic, lipid lowering, or antihypertensive drugs, and information on the hospital such as quartile of procedure volume, geographic location (North-West, South-West, East Germany), and hospital size using a quotient of procedure volume divided by the total number of patients treated. Additionally, models were adjusted for number of endovascular or open surgical procedures during stay, length of stay, number of days with surgeries, emergency department admission, and number of affected lesion levels and legs. In addition to these variables, comorbidities were selected from Elixhauser comorbidity groups by elastic nets and varied between models.
Each model was amended by variables based on clinical relevance. The model for in hospital mortality also contained the van Walraven score that summarises the comorbidity burden into a weighted sum score based on the Elixhauser groups.
The adjusted variables are displayed in Supplementary Table S2. The alpha level was set at 5%. Statistical analysis was performed using SAS 9.4 (SAS Institute, North Carolina, USA) and R version 4.0.3 (The R Foundation for Statistical Computing, Vienna, Austria). A list of R packages that were used can be found in Supplementary Table S3.
Results
The study cohort consisted of 88 187 cases (55.9% intermittent claudication) comprising 59 066 individual patients who underwent invasive revascularisation in 668 hospitals (Figure 1); the mean age was 72.5 years and 44.3% were female. Follow up was censored at one year, resulting in a median follow up of 365 days (mean 336 days). The data were stratified by procedure type, resulting in 63 883 cases with EVR and 24 304 with OSR (Table 1). For analyses of freedom from major amputation, patients with previous major amputations were excluded, leading to a sample of 36 113 EVR and 13 520 OSR cases (Fig. 1).
Figure 1Flowchart of inclusion criteria for selection of patients aged ≥ 40 years with symptomatic peripheral artery disease, treated with endovascular or open surgical revascularisation in Germany between 2013 – 2018.
Table 1Characteristics of the study population of 59 066 patients with symptomatic peripheral arterial disease of the lower limbs, shown separately by endovascular or open surgical revascularisation
Characteristics
Endovascular procedure
Open surgical procedure
Hospital cases – n
63 883
24 304
Individual patients – n
44 725
20 146
Hospitals – n
640
571
Female sex
28 864 (45.29)
10 192 (41.9)
Age – y
72.70 ± 10.30
71.87 ± 10.05
Fontaine stage 2
37 765 (59.1)
11 550 (47.5)
Fontaine stage 3
6 324 (9.9)
4 775 (19.6)
Fontaine stage 4
19 794 (31.0)
7 979 (32.8)
Lipid lowering medication
38 235 (59.9)
14 185 (58.4)
Antithrombotic medication
40 214 (62.9)
15 057 (62.0)
Antihypertensive medication
56 056 (87.7)
20 975 (86.3)
Emergency department admission
11 362 (17.8)
6 009 (24.7)
Procedures involving both legs
5 759 (9.0)
1 778 (7.3)
Treatment of lesion below the knee
21 349 (33.4)
6 724 (27.7)
Number of lesion sites, right/left leg and three lesion levels
1.32 ± 0.54
1.43 ± 0.62
Median van Walraven Score
10 [3, 19]
11 [4, 20]
Data are presented as n (%), mean ± standard deviation, or median [interquartile range] unless stated otherwise.
A minority of cases involved procedures performed on both legs (9.0% for EVR and 7.3% for OSR) and fewer patients were revascularised below the knee (33.4% for EVR and 27.7% for OSR). Patients who underwent OSR presented with more comorbidities and more often with chronic limb threatening ischaemia (52.4%) when compared with EVR. Baseline characteristics stratified by procedure type and hospital volume quartiles can be found in Table 2 and Table 3. Information on frequency of outcome events is given in Supplementary Table S4. Additional sensitivity analyses are given in Supplementary Table S5.
Table 2Baseline characteristics of the included 59 066 patients with symptomatic peripheral arterial disease of lower limbs who underwent endovascular revascularisation for symptomatic peripheral arterial disease by hospital procedure volume in quartiles
Variable
1st Quartile
2nd Quartile
3rd Quartile
4th Quartile
Hospitals – n
483
228
137
59
Patients – n
16 248
12 919
10 811
8 695
Procedures
236 [124, 385]
748 [650, 869]
1 279 [1 128, 1 481]
2 286 [1 993, 2 755]
Female sex
9 369 (45.4)
7 639 (45.7)
6 515 (45.4)
5 341 (43.8)
Age – y
72.66 ± 10.40
72.72 ± 10.27
72.88 ± 10.30
72.53 ± 10.17
Fontaine stage 2
11 936 (57.8)
10 127 (60.6)
8 229 (57.4)
7 473 (61.3)
Fontaine stage 3
2 075 (10.0)
1 560 (9.3)
1 446 (10.1)
1 243 (10.2)
Fontaine stage 4
6 648 (32.2)
5 018 (30.0)
4 660 (32.5)
3 468 (28.5)
Lesion
Aorto-iliac
5 243 (25.4)
4 418 (26.4)
3 509 (24.5)
3 187 (26.2)
Femoropopliteal
12 888 (62.4)
10 776 (64.5)
9 425 (65.7)
8 380 (68.8)
Crural
6 801 (32.9)
5 326 (31.9)
5 130 (35.8)
4 092 (33.6)
Van Walraven score
10.00 [3.00, 18.00]
10.00 [3.00, 18.00]
11.00 [4.00, 19.00]
11.00 [4.00, 18.00]
Congestive heart failure
7 367 (35.7)
6 037 (36.1)
5 267 (36.7)
4 232 (34.7)
Cardiac dysrhythmias
8 276 (40.1)
6 690 (40.0)
6 080 (42.4)
4 931 (40.5)
Hypertension
19 081 (92.4)
15 527 (92.9)
13 386 (93.4)
11 339 (93.1)
Chronic pulmonary disease
9 002 (43.6)
7 350 (44.0)
6 395 (44.6)
5 440 (44.6)
Uncomplicated diabetes
10 322 (50.0)
8 307 (49.7)
7 291 (50.9)
5 986 (49.1)
Complicated diabetes
8 759 (42.4)
6 965 (41.7)
6 177 (43.1)
5 077 (41.7)
Renal failure
7 884 (38.2)
6 727 (40.3)
6 307 (44.0)
5 320 (43.7)
Liver disease
4 725 (22.9)
3 687 (22.1)
3 300 (23.0)
2 815 (23.1)
Obesity
5 867 (28.4)
4 810 (28.8)
4 230 (29.5)
3 401 (27.9)
Prior stroke or TIA
3 237 (15.7)
2 473 (14.8)
2 331 (16.3)
1 884 (15.5)
History of coronary artery disease
10 751 (52.0)
8 764 (52.5)
7 681 (53.6)
6 576 (54.0)
Prior myocardial infarction
2 407 (11.7)
1 961 (11.7)
1 826 (12.7)
1 544 (12.7)
Dyslipidaemia
16 146 (78.2)
13 427 (80.4)
11 675 (81.4)
10 229 (84.0)
Antihypertensives in previous year
18 009 (87.2)
14 710 (88.1)
12 644 (88.2)
10 693 (87.8)
Lipid lowering drugs in previous year
11 800 (57.1)
9 824 (58.8)
8 662 (60.4)
7 949 (65.2)
Antithrombotic drugs in previous year
12 585 (60.9)
10 375 (62.1)
9 254 (64.6)
8 000 (65.7)
Emergency department admission
3 797 (18.4)
3 215 (19.2)
2 459 (17.2)
1 891 (15.5)
Number of prior medications
10.00 [6.00, 15.00]
10.00 [7.00, 15.00]
11.00 [7.00, 15.00]
10.00 [7.00, 15.00]
Data are presented as n (%), mean ± standard deviation, or median [interquartile range] unless otherwise stated. TIA = transient ischaemic attack.
Table 3Baseline characteristics of the included 59 066 patients with symptomatic peripheral arterial disease of the lower limbs who underwent open surgical revascularisation for symptomatic peripheral arterial disease by hospital procedure volume in quartiles
Variable
1st Quartile
2nd Quartile
3rd Quartile
4th Quartile
Hospitals – n
391
211
150
76
Patients – n
7 920
5 797
5 414
5 173
Procedures
207 [116, 312]
566 [499, 631]
839 [767, 926]
1 265 [1 134, 1 581]
Female sex
3 180 (40.2)
2 368 (40.8)
2 339 (43.2)
2 305 (44.6)
Age – y
71.70 ± 10.13
71.89 ± 10.01
72.01 ± 9.99
71.95 ± 10.04
Fontaine stage 2
3 745 (47.3)
2 666 (46.0)
2 599 (48.0)
2 540 (49.1)
Fontaine stage 3
1 543 (19.5)
1 161 (20.0)
1067 (19.7)
1 004 (19.4)
Fontaine stage 4
2 632 (33.2)
1 970 (34.0)
1 748 (32.3)
1 629 (31.5)
Lesion
Aorto-iliac
1 621 (20.5)
1 313 (22.6)
1 499 (2.7)
1 536 (29.7)
Femoropopliteal
6 542 (82.6)
4 814 (83.0)
4 478 (82.7)
4 416 (85.4)
Crural
2 091 (26.4)
1 593 (27.5)
1 553 (28.7)
1 487 (28.7)
Van Walraven score
11.00 [4.00, 19.00]
11.00 [4.00, 19.00]
11.00 [4.00, 19.00]
12.00 [5.00, 20.00]
Congestive heart failure
2 853 (36.0)
2 123 (36.7)
1 915 (35.4)
1 923 (37.2)
Cardiac dysrhythmias
3 128 (39.5)
2 317 (40.0)
2 117 (39.1)
2 115 (40.9)
Hypertension
7 272 (91.8)
5 341 (92.1)
5 016 (92.6)
4 727 (91.4)
Chronic pulmonary disease
3 660 (46.2)
2 757 (47.6)
2 523 (46.6)
2 515 (48.6)
Uncomplicated diabetes
3 658 (46.2)
2 622 (45.2)
2 344 (43.3)
2 240 (43.4)
Complicated diabetes
2 950 (37.2)
2 162 (37.3)
1 944 (35.9)
1 861 (36.0)
Renal failure
2 740 (34.6)
2 064 (35.6)
2 000 (36.9)
1 987 (38.4)
Liver disease
1 852 (23.4)
1 319 (22.8)
1 252 (23.1)
1 237 (23.9)
Obesity
2 068 (26.1)
1 463 (25.2)
1 363 (25.2)
1 355 (26.2)
Prior stroke or TIA
1 202 (15.2)
872 (15.0)
811 (15.0)
762 (14.7)
History of coronary artery disease
4 122 (52.0)
3 096 (53.4)
2 849 (52.6)
2 753 (53.2)
Prior myocardial infarction
995 (12.5)
826 (14.2)
663 (12.2)
665 (12.8)
Dyslipidaemia
6 134 (77.4)
4 510 (77.8)
4 247 (78.4)
4 131 (79.9)
Antihypertensives in previous year
6 841 (86.4)
4 999 (86.2)
4 703 (86.9)
4 432 (85.7)
Lipid lowering drugs in previous year
4 570 (57.7)
3 465 (59.8)
3 183 (58.8)
2 967 (57.4)
Antithrombotic drugs in previous year
4 892 (61.8)
3 639 (62.8)
3 341 (61.7)
3185 (61.6)
Emergency department admission
2 008 (25.4)
1 400 (24.2)
1 296 (23.9)
1 305 (25.2)
Number of prior medications
10.00 [6.00, 14.00]
10.00 [6.00, 15.00]
10.00 [6.00, 14.00]
10.00 [6.00, 15.00]
Data are presented as n (%), mean ± standard deviation, or median [interquartile range] unless otherwise stated. TIA = transient ischaemic attack.
Analyses of re-interventions did not show a significant association with hospital procedure volume for both procedure types. The risk of major amputation was statistically significantly lower after a procedure in a high volume hospital for both procedure types with an HR of 0.72 (95% CI 0.52 – 0.99) for EVR and 0.61 (95% CI 0.44 – 0.85) for OSR. Patients treated by OSR in high volume hospitals enjoyed better survival (HR 0.85; 95% CI 0.73 – 0.98) when compared with the reference. No statistically significant difference in midterm survival was found for patients with EVR (Figure 2, Figure 3).
Figure 2Outcomes after 63 883 endovascular procedures for patients with symptomatic peripheral arterial disease of the lower limbs, shown separately for each quartile of annual hospital volume (reference: first quartile), adjusted for patient and procedure characteristics, medication, comorbidities, and prior treatment in Germany between 2013 – 2018. HR = hazard ratio; OR = odds ratio.
Figure 3Outcomes after 24 304 open surgical procedures for patients with symptomatic lower limb peripheral arterial disease, shown separately for each quartile of annual hospital volume (reference: first quartile), adjusted for patient and procedure characteristics, medication, comorbidities, and prior treatment in Germany between 2013 – 2018. HR = hazard ratio; OR = odds ratio.
For EVR patients, the second quartile showed a statistically significant increase of in hospital mortality compared with the first quartile (OR 1.28; 95% CI 1.03 – 1.58) (Figure 2, Figure 3). Sensitivity analyses were confirmative (Supplementary Table S3).
Discussion
In this analysis of insurance claims from Germany, data were included on almost 65 000 individuals who underwent more than 88 000 hospitalisations for the invasive treatment of symptomatic PAD between 2013 and 2018. For the first time, it linked available statistical data on annual procedure volume with patient related midterm outcomes to determine the complex relationship between volume and outcome. It adjusted for patient and procedure characteristics, medication, comorbidities, and prior inpatient and outpatient information, while all analyses were stratified by the index approach. Interestingly, patients who underwent both EVR and OSR in a high volume centre were at lower risk of major amputation in the longer term, while an additional survival benefit was observed during 12 months after OSR in high volume centres.
Few articles have addressed the aspects of case or procedure volume in relation to outcomes after invasive treatment of PAD. In two recently published articles by Iida et al., the authors used data from multicentre registries in Japan.
Impact of institutional volume on critical in-hospital complications adjusted for patient- and limb-related characteristics: an analysis of a nationwide Japanese registry of endovascular interventions for PAD.
While 0.3% of the nationwide cohort encountered critical in hospital complications after EVR for symptomatic PAD, a higher institutional volume was statistically significantly associated with a lower risk of events.
Impact of institutional volume on critical in-hospital complications adjusted for patient- and limb-related characteristics: an analysis of a nationwide Japanese registry of endovascular interventions for PAD.
A propensity score matched analysis of 236 pairs revealed that lower institutional volume was associated with higher 12 month re-stenosis rate after aorto-iliac stenting for PAD, while comparable in hospital outcomes were observed.
The study design and cohorts widely differed. Kantonen et al. found that surgeon’s caseload and hospital volume affected amputation rates but not the mortality rate in patients operated on for chronic limb threatening ischaemia.
In contrast, Troeng et al. reported that experience of the surgeon and the type of hospital were less important after an operation for chronic leg ischaemia.
Using data from the Oxford region, Michaels et al. concluded that districts with high rates of distal arterial reconstructions performed fewer amputations and that more widespread use of reconstructive surgery would result in substantial reductions in the number of leg amputations.
Bates et al. had access to data from the Veterans Affairs programmes; their study did not show an association between higher amputation case volume and short term mortality, while no data on revascularisation procedures were available.
While the latter studies included rather small cohorts, Moxey et al. used large hospital episode statistics data from the UK comprising a total of 27 660 femoropopliteal bypass and 4 161 femorodistal bypass procedures. A positive volume–outcome relationship was found for these procedures with benefits in terms of mortality and limb salvage both during the short and midterm follow up.
In an approach to summarise the available evidence, Awopetu et al. conducted a systematic review and concluded that higher volume hospitals were associated with reduced amputation and mortality rates after lower limb vascular surgery; however, the authors also emphasised that these data were not conclusive due to significant heterogeneity.
Besides the fact that all these previous studies applied heterogeneous study designs, they partially included historical cohorts while the field of PAD treatment rapidly developed during the past two decades.
More recently, Scali et al. used data from the Society for Vascular Surgery Vascular Quality Initiative registry comprising a total of 25 852 procedures with sufficient one year follow up (2003 – 2019). A higher volume was associated with diminished midterm survival, while increasing practice experience in years was more significantly associated with a reduction of in hospital complications and the risk of major adverse limb events compared with the volume. In contrast, neither experience nor volume had any significant association with early mortality.
Interestingly, the results from the current study overall confirmed the inconsistency in previous studies. The rather distinct benefits in terms of major amputation rate were accompanied by a survival benefit only after open surgical procedures.
Strengths
This study explored the role of hospital volume in the context of PAD in Germany based on an innovative linkage of data extracted from hospital quality reports and insurance claims data. A particular strength was the longitudinal design, which enabled a current or history of chronic diseases and prior treatment to be taken into account, and also outcomes occurring beyond discharge to be considered.
Limitations
Although the data offer a valuable insight into the effect of hospital volume on patient outcomes, they have some limitations. The administrative data used for the current study were not primarily collected for research purposes and it cannot be excluded that systematic coding errors appeared. Extensive data cleaning involving internal validation and Z scores with median absolute deviation was performed. Furthermore, the issue of residual confounding in observational research remains unsolved. Hence it seems important to question clinical relevance against statistical significance. Despite the large sample size of the study, in hospital death is a rare event after lower extremity revascularisation, which therefore leads to broad confidence intervals. Lastly, the complex association between hospital volumes, surgeons’ teaching curriculum, who treats patients in an academic centre, and other aspects certainly deserve more attention, as the available information were limited in the current study. To date, 36 university medical centres exist beside more than 1 600 general hospitals in Germany. If a non-academic hospital is registered as a teaching hospital, it does not follow commonly accepted assumptions concerning surgical training and there is no central registry for this information.
The results of the current study emphasise that the vascular community should further explore the possible relationship between case volume and outcomes in the PAD field. In times of increasing centralisation of procedures in vascular surgery, it appears reasonable to consider all the different areas of clinical expertise.
Conclusions
This large analysis of health insurance claims revealed that a higher case volume was associated with lower major amputation rate in the longer term after both endovascular and open surgical revascularisation for symptomatic PAD. Furthermore, midterm survival was better after open surgical revascularisation in high volume facilities. These results confirm previous data emphasising that the vascular community should expand considerations of centralisation to domains other than aortic repair. Beyond the pressing need for a standardisation of definition of volume stratification, future projects should aim to derive appropriate thresholds for the treatment of patients with PAD.
Conflicts of interest
None.
Acknowledgements
The authors are grateful for the scientific support from BARMER. This work was supported by the German Federal Joint Committee (grant number 01VSF18035 to CAB).
Appendix A. Supplementary data
The following is the Supplementary data to this article:
Editor's Choice - The impact of centralisation and endovascular aneurysm repair on treatment of ruptured abdominal aortic aneurysms based on international registries.
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.
Long term evaluation should be an integral part of the clinical implementation of new vascular treatments - an ESVS Executive Committee position statement.
Editor's Choice - International variations and sex disparities in the treatment of peripheral arterial occlusive disease: a report from VASCUNET and the International Consortium of Vascular Registries.
Impact of institutional volume on critical in-hospital complications adjusted for patient- and limb-related characteristics: an analysis of a nationwide Japanese registry of endovascular interventions for PAD.
Editor's Choice - 2017 ESC Guidelines on the Diagnosis and Treatment of Peripheral Arterial Diseases, in collaboration with the European Society for Vascular Surgery (ESVS).
Hospital volume and mortality for 25 types of inpatient treatment in German hospitals: observational study using complete national data from 2009 to 2014.
Editor's Choice - long term survival after femoropopliteal artery revascularisation with paclitaxel coated devices: a propensity score matched cohort analysis.
A retrospective cohort study on the provision and outcomes of pharmacological therapy after revascularization for peripheral arterial occlusive disease: A study protocol.
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.
00808-5&id=gr1.jpg){kind=link}
00808-5&id=gr2.jpg){kind=link}
00808-5&id=gr3.jpg){kind=link}