DATA COLLECTION FOR PCI: HOW IT WORKS

DATA COLLECTION FOR PCI: HOW IT WORKS

Section detailing the mechanisms in place for collection of PCI data.

The original BCIS-CCAD dataset for PCI was introduced in 2002 and has since been updated on several occasions.  Details are on the ‘BCIS-CCAD datasets and history’ page. There are several ways in which units can collect these data.

 

BASIC SET-UP

Data are entered via the NCAP section of the NICOR web site.  There are 2 options:

1.  A local database can be used to for data entry.  This database will need to create an export file of the specified csv format, and this can then by uploaded to the NICOR website.

2.  Direct data entry.  Log in to the NCAP section of the NICOR web site, log into the PCI section and enter procedure specific data into the web forms provided

 

1. DIRECT DATA ENTRY

2. DATA ENTRY VIA DEDICATED DATABASE PROGRAM:

2.1 BCIS PCI DATABASE

I have designed a PCI database program using Microsoft Access. This is available for free download on the ‘Current PCI database’ page. In addition to providing procedure reports and a variety of analyses, it creates a commas separated values (csv) file in the format required for upload to NICOR.

2.2 COMMERCIAL DATABASES

There are a number of commercially available database systems. These also all work by extracting the required data to create a csv file in the appropriate format for upload to NICOR.

3. WEB BROWSER

Peter Ludman 2020

BCIS Peer Review Service

BCIS Peer Review Service

BCIS are pleased to provide its Peer Review Service service, designed to assist researchers and available without charge to all BCIS members.

OUR AIM:

To help BCIS members, who have developed a research proposal, to further refine this through constructive review and feedback – thus preparing researchers to navigate ‘hurdles’ such ethics committees, funding applications etc. and progress to active projects with a high likelihood of successful completion and reporting.

WHO CAN APPLY?:

This service will be available to any current BCIS member, irrespective of their current post, or previous research experience.

WHAT WE CAN DO FOR YOU:

The peer-review service is able to provide the following:

Principal Review

Review, by 2-3 consultant-level clinicians or senior academics, of the research proposal. Reviewers will usually be drawn from the BCIS membership, although for specific issues where other expertise is needed, external review may sometimes be sought.

We aim to provide a rapid turnaround, with these responses usually fed back to the submitting member within 4 weeks.

Additional Specialist Review

Applicants may request additional specialist review in the following domain:

i) Specialist statistical review – in specific cases where complex statistical methods are necessary as part of the protocol and where expertise in the relevant area is not available within BCIS

The BCIS research group will seek to commission these services for selected projects that are at an advanced stage of development and that are deemed, from the main review, to be close to finalisation.

Applicants not selected for immediate specialist review may re-apply as their project matures, for example after resolution of key issues raised in the principal review.

KEY GUIDING PRINCIPLES UNDERLYING THE BCIS PEER REVIEW SERVICE:

CONFIDENTIALITY – including the following working rules:

i) Proposals will only be seen by, and circulated amongst, the selected reviewers and the peer-review scheme coordinators

ii) Feedback will be provided directly via the coordinators to the submitting researcher only

iii) The option to choose anonymised (i.e. blind) review, if the submitting member prefers

KEY GUIDING PRINCIPLES UNDERLYING THE BCIS PEER REVIEW SERVICE:
CONSTRUCTIVE – hence no reason for members who are considering making a submission for peer review to feel intimidated about seeking BCIS input.

COLLEGIATE – ownership of the idea and the project will remain entirely with the submitting researcher.

WHAT TO DO NEXT

1) Please download and read the two “Getting Started” documents below.

Getting Started One.

Getting Started Two.

– Getting Started 1 is a description of the anticipated format and content of research proposals to be submitted to BCIS for peer review.

– Getting Started 2 is a checklist of further specific details that we will need, in order to provide the most useful research feedback possible to users of the service.

2) Once you are ready to submit for peer review, please email the research protocol, together with the completed checklist, to the email address below:
bcis@millbrookconferences.co.uk

BCIS Research and Development Group, January 2015

CURRENT PCI DATABASE

CURRENT PCI DATABASE

A database designed to capture details about percutaneous coronary intervention procedures.

This PCI Database is designed to capture details about percutaneous coronary interventional procedures. Procedure reports and GP letters can be created, the data can be analysed within the database, or exported for analysis. This database is designed to collect the dataset agreed by the British Cardiovascular Intervention Society (BCIS) / National Institute for Cardiovascular Outcomes Research (NICOR). It can create the comma separated values (csv) file required for upload data to the NICOR servers. I have written the database in Access 2002.  Version 19.0 is the latest and incorporates library updates (as 2021)

 

Jan 2021

Version 19.0

The latest version of the database below is version 19.0.  I suggest that anyone who wants to start from scratch using this database starts with version 19.0  The three files required are available for download here:

  1. PCI Database Version 19.0
  2. Mousehook dll
  3. Release Notes for version 18.x

 

 

For earlier versions of the database:

May 2019 Menu Options update

To incorporate the new library options introduced in May 2019, 4 tables in the database need to be replaced with new versions.  Instructions to do this are in this file.

The file containing those tables can be downloaded at this link (it will need to be unzipped once downloaded) BCIS database updated tables

This database is not locked so the code can be explored and modified if necessary (see Release Notes ‘getting inside the database’). You can ask your IT department to set up a link between this database and your local hospital Patient Administration System, so that patient demographics, NHS numbers etc. do not have to be entered manually. A link to the lab results server is also very helpful.

The database is designed so that most of the data is entered during and at the end of a PCI procedure, and a procedure report can be printed for the patient notes at that stage. A final set of data is entered after patient discharge.

The data can be analysed using the queries I have written, and in addition the full dataset (excluding sub-tables) can be exported to Microsoft Excel to be analysed elsewhere. In addition once the data are uploaded to NICOR (via the NCAP web site), there is a suite of tools to analyse the live data, with National comparisons available.

 

By Peter F Ludman, BCIS Audit Lead

RELATED RESOURCES

BCIS PCI DATA COLLECTION AND VALIDATION: HOW TO AND GUIDANCE DOCUMENTS

BCIS PCI DATA COLLECTION AND VALIDATION: HOW TO AND GUIDANCE DOCUMENTS

Downloadable guidance documents regarding the upload of PCI data to NICOR and analysis to produce the Clinical Outcomes Publication validation and final report

RELATED RESOURCES

BCIS PCI DATASETS – UPDATES HISTORY

BCIS PCI DATASETS – UPDATES HISTORY

CURRENT DATASET AND VERSION

 

16th May 2019

To all database PCI operators, database contacts and database software providers

Dear colleagues,

I have updated the PCI dataset libraries, to try to capture more recent changes to our armamentarium. These are ONLY library updates, and affect only 4 libraries.  There is no change to the overall structure of the database.

We would hope that most centres will be able to update their library options rapidly, but would hope that all will have done this by the next 6 months at the latest.

The dataset is now named: ‘PCI_Dataset_Standard 5_6_5 version 1-05-2019’ and can be downloaded as an excel file from:

 

The changes are summarised below:

5.09       ECG ischaemia

The following have been added:

  1. CT fractional flow reserve
  2. QFR (Quantitative Flow Ratio or equivalent angiographic analysis of flow)

3.17 Drug eluting stents

The following have been added:

  1. Papyrus (Biotronik)
  2. Graft master (Abbott)
  3. Over and Under (ITGI)
  4. BeGraft (Bentley)
  5. Magmaris (Biotronik)
  6. EluNIR (Cordis Cardinal Health)
  7. Xience Alpine (Abbott)
  8. Xience Sierra (Abbott)
  9. Supraflex (SMT)
  10. Supraflex Cruz (SMT)
  11. Supralimus (SMT)

3.19 Diagnostic device(s) used during procedure

  1. Pressure wire

Will now be used for any device that assesses a coronary stenosis and requires hyperaemia to be induced. It has been re-named:

  1. Pressure wire (or any other HYPERAEMIC based index of coronary stenosis)

The following have been added:

  1. iFR, RFR, dPR (or any other NON HYPERAEMIC index of coronary stenosis)
  2. NIRS (Near infra-red spectroscopy)

3.20 Procedural device:

  1. Rotational atherectomy

Will now be used for the legacy device and so be renamed

  1. Rotational atherectomy (Rotablator legacy device)
  2. Finecross

Will now be used for both Finecross and ANY other microcather and so has been renamed:

  1. Finecross (or any other microcatheter not in list)

The following have been added:

  1. Drug eluting balloon
  2. Shockwave balloon
  3. Coronary sinus occlusion device – PICSO (Miracor)
  4. Rotational atherectomy (Rotapro device)

Please note  that where options have been re-named, the previous and new names will both be accepted

There is no change to the underlying structure of the dataset.

The only thing you need to do is to update your library drop down lists.

4th August 2016

Dataset Version 5.6.5 (August 2016)

Sent to all commercial DB vendors and all Database audit contact and PCI database clinical leads:

I have updated the BCIS PCI dataset. The ‘change history’ tab in the spreadsheet gives the details. They are as follows:
1. A text clarification
2. Clarification of an endpoint definition
3. Minor additions to three of the libraries as:

Library update: 3.17 Drug eluting stent – Addition 46. Orsiro (Biotronik)

Library update: 3.21 Athero-thrombus removal device(s) used – Addition: 13. QXT extraction catheter

Library update: 5.10 Drug therapy PreOp – Addition: 10. Cangrelor

There is no change to the underlying structure of the dataset.

The only thing you need to do is to update your library drop down lists.


30th October 2014

Sent to all All commercial DB vendors and All Database audit contacts

Dataset Version 5.6.2 To be collected from 1st December 2014

Addition to the library of Drug Eluting Stent names only. No change to the database structure.


29th October 2013
Sent to: All commercial DB vendors, All UK PCI operators, All Database audit contacts

Dataset version 5.6.1 To be collected from 1st January 2014

I am always reluctant to change our dataset, but unfortunately there are compelling reasons to make some small changes. There are 11 additional fields (6.01 to 6.11) and additional options to an existing field.

Eleven additional fields:
6.01: PCI indication for stent thrombosis. It has previously been impossible to identify these cases as they could fall under several options available in the ‘Indication for PCI’ field. This new field will allow us to track this particular indication for PCI for the first time.

6.02: As we move towards international standardisation of definitions, we felt it important to have uniform reporting of bleeding complications. This field aligns us to the Bleeding Academic Research Consortium (BARC) definitions.

6.03 to 6.11: Out Of Hospital Cardiac Arrest. We have only been able to infer this specific patient presentation by using the field ‘need for pre-procedural ventilation’. As out of hospital arrest carries its own very specific high risk features we felt it critical to try to capture some key features of patients presenting in this way for PCI. I hope that the data items selected provide the correct balance between essential detail without too burdensome data entry.

Update to existing fields:
3.24 Circulatory support. New LV support device potion have been added to include: Impella, Lucas device, Autopulse and ECMO

What you need to do
1. The commercial suppliers of PCI databases are all being informed of these modifications so that their systems can be upgraded to remain compliant with the BCIS – NICOR audit collection. I recommend you contact then to make sure changes can be effected in time.

2. The Lotus Notes database will be updated in the near future.

3. If you are using my PCI database (in access), then I will shortly be providing a method to allow you to update your database to a newer version that will incorporate all these changes

Best wishes

Peter Ludman
BCIS audit lead


3 September 2013
Sent to all BCIS members and database contacts

Diagnostic Interventional Procedures
The use of FFR, IVUS and OCT (and OFDI etc) when no PCI ensues will be described as an ‘Interventional Diagnostic Procedure’.

An Interventional Diagnostic Procedure is therefore defined as invasive coronary angiography with the use of adjunctive invasive diagnostic equipment such that a coronary device approaches, probes or crosses one or more coronary lesions (including – but not limited to – a pressure wire, intravascular ultrasound and swept laser imaging), before the intention to treat by mechanical revascularisation has been decided. Interventional diagnostic cases should be performed by interventional cardiologists in intervention capable centres.

The definition of a PCI will remain unchanged as:
“A percutaneous coronary intervention (PCI) is deemed to have taken place if any coronary device approaches, probes or crosses one or more coronary lesions, with the intention of performing a coronary intervention. Usually this device will be a guide wire. The only exception to this will be patients who have an adverse cardiac event (during an attempted PCI) that necessitates procedure termination prior to the introduction of a coronary device. This rare type of case will also be defined as a PCI and therefore this will classify as a complication.”

BCIS will plan to report operator’s total numbers of ‘PCI procedures’ and, separately, total numbers of ‘Interventional Diagnostic Procedures’

In order to do this all ‘Interventional Diagnostic procedures’ should now be entered into the BCIS database. To distinguish these from PCI procedures, the following rules need to be applied:

For an ‘Interventional Diagnostic Procedure’:
Total number of lesions attempted (field 3.11) must = 0
and Total number of vessels attempted (field 3.10) must =0
and ‘Diagnostic device(s) used during procedure (field 3.19) should include one of the interventional diagnostic procedures described
Vessels attempted (field 3.09) should be left empty
(Any adverse in-hospital outcomes should be recorded in the same way as for a PCI procedure)

It is suggested that recording all ‘Interventional Diagnostic Procedures’ should start from Jan 2014, but those wishing to add data retrospectively can do so, and it may be possible to include this in the 2013 data reports to be produced in 2014.
Peter Ludman


April 2013
Sent to: All commercial DB vendors, All BCIS members, All Database audit contacts

Dear all

Dataset version 5.5.6 to 5.5.7: To be applied by 1st June 2013
No structural change to the database.
The only changes are to update of some of the menu items in the volatile fields, and some clarification of descriptors.

1. Volatile fields
The following additions have been made to the list:
3.17 Drug based stents

31 Synergy (Boston Scientific)
32 Promus premier (Boston Scientific)
33 Biofreedom (Biosensors)
34 Combo (Orbus)
35 Mguard (InspireMD)
36 Xience Pro (Abbott)
37 Absorb BVS (Abbott)

2. Dataset clarification
Minor changes to clarify the meaning of items:
5.09 ECG Ischaemia. Option 3 ‘on perfusion scan’. This should also be selected for other stress imaging such as stress echo and stress cardiac magnetic resonance imaging.
3.19 Diagnostic device(s) used during procedure. Option 6 relabelled ‘OCT (Optical Coherence Tomography) or similar’. This should be selected for all similar laser swept intracoronary imaging techniques such as optical frequency domain imaging (OFDI).

Attached files:
1. A full history of the dataset changes are documented (Dataset version history)
2. A complete spreadsheet of the new version 5.5.7
(The csv file specification is unchanged)
All these will also be available on the BCIS web site

What you need to do
1. The commercial vendors of databases are all being informed of these modifications.

2. The Lotus Notes database will be updated to include these shortly.

3. If you are using my PCI database, then these changes can be added to the lists by going to the ‘default settings / field lists / database admin’ button on the main switchboard, and then selecting each list to be update in turn and typing in the additions.

Best wishes

Peter Ludman
BCIS Audit Lead
peter.ludman@uhb.nhs.uk


October 2012
Import system and rules for BCIS-CCAD 2012 version

The logical and internal validation rules used by NICOR to check for data consistency during all data uploads can be found here.

Peter Ludman
BCIS Audit Lead
peter.ludman@uhb.nhs.uk


Risk Adjusted Funnel Plots and Delays Report

December 2008

The risk adjusted outcome cummulative funnel plot generator for PCI is now ready to be used. The report works in the same way as the generation of aggregate data reports, and the time delays reports generator. Via the Lotus Notes program in BCIS section, look under the list of reports on the left of the page. Click on ‘ Funnel Plot’. Work your way through the steps, which represent a series of choices regarding the variables and parameters you would like to use to analyse your data .

The report will generate an analysis in the form of a password protected excel spreadsheet. The predicted and observed MACCE are listed for each PCI procedure using your reported CCAD data and the NQWIP model. A cummulative funnel plot is generated with 2 and 3 sigma lines. You will find an explanation of this plot in the last few slides (number 156 to 177) of my BCIS audit presentation of 2007 data at BCIS Audit Returns 2007.

The latest version of the delays report generator for PCI is now ready to be used. The report works in the same way as the generation of aggregate data reports. Via the Lotus Notes program in BCIS section, look under the list of reports on the left of the page. Click on ‘Delay reports v.6’ This will bring up the following dialogue (please see attachment for dialogue).

Work your way through the 7 steps, which represent a series of choices regarding the variables and parameters you would like to use to analyse your data.

The report will generate an analysis in the form of a password protected excel spreadsheet. The worksheets include a measure of your data completeness for the key fields, and explanation of how each field is calculated, and overall analysis of various time intervals in the treatment of patients being treated for acute coronary syndromes. The report will aslo create statistical process control charts allowing you to see (for example) each individual door to balloon time delay for each patient treated for STEMI. These charts can be split to look at night versus day time delays.

I hope you find this is helpful. If you spot any errors in our programming please do not hesitate to contact me.

Peter Ludman
BCIS Audit Lead

RELATED RESOURCES