Johns Hopkins Hospital aka John Hopkins Hospital

Bayview Medical Center aka Francis Scott Key Hospital Baltimore City Hospital

~20 STEMI / yearPrimary PCI

~120 STEMI / yearThrombolytics Only

451 Patient 11 Community Hospitals

Without On-Site Cardiac Surgery

Catheterization & PTCA225

Thrombolytic Therapy226

Composite 6 week & 6 Month EndpointDeath, Non-fatal MI, Stroke

Composite 6 week & 6 Month EndpointDeath, Non-fatal MI, Stroke

Outcomes at 6 WeeksP=0.03

C-PORT RegistryMajor Adverse Events

C-PORT RegistryOutcomes in Context

Cannon, et. al. JAMA 2000; 283:2941

Magrid, et. al. JAMA 2000;284:3131

Keeley, et. al. Lancet. 2003;361(9351):13-20

ASSENT Investigators Lancet. 2001;358:605-13

CPORT- E TrialRandomized trial comparing medical, economic and

quality of life outcomes of non-primary PCI at hospitals

with and without on-site cardiac surgery

Motivation for Trial

Sustain primary PCI program at no-SOS hospitals

Improve access to PCI servicesReduce pressure to create additional cardiac

surgery programsNeed for research to inform healthcare policy

decisions by state and national organizations

Approving a medical experiment that increases mortality 38% What were those N. J. bureaucrats thinking ?

Wennberg DE, et. al. JAMA. 2004 Oct 27;292(16):1961-8.

Wennberg DE, et. al. JAMA. 2004 Oct 27;292(16):1961-8

Patient for Diagnostic Cath

Informed consent

Catheterization

Meetsinclusioncriteria

PCI no SOS PCI with SOS

Exclusioncriteria

Consent Registry

Refuse

NoConsent Registry

Not Approached

3:1 Randomization

Analysis by intention-to-treat

Study EndpointsNon-inferiority trial

Primary Endpoints

All-cause mortality at 6 weeks

MACE at 9 months All-cause mortality Q-wave myocardial infarction Target vessel revascularization

Assuming 6 week mortality to be 0.8% and 9 month MACE to be 12%, a sample size of 18360 was selected to define a non-inferiority margin of 0.4% for mortality and 1.8% for MACE with a one-sided test for non-inferiority using α=0.05 and β=0.80 for both primary endpoints.

Inclusion and Exclusion CriteriaPatient InstitutionInclusion• Age > 18 years• Informed

consent• > 50% stenosis• All target lesions

approachable at no-SOS hospital

Exclusion• Unprotected LM• EF < 20%• MD-judged high

risk

DevicesInclusion• Balloon, stent• Distal

protection• Covered stent• Cutting balloon

- in-stent restenosis

Exclusion• Atherectomy • Cutting

balloon- de novo lesion

Inclusion• > 200

PCI/year• 24/7 Primary

PCI• Complete

formal development program

• Interventionalist meets AHA/ACC competency

Participating Centers

60 Centers

Center Annual PCI Procedure Volume

150 (99,216) median (25th,75th percentile)

Median Participation Duration 2.2 years

No-SOS SOS p-valueAge (years) (mean+/-SD) 64+/-12 64+/-12 0.42

Male Gender (%) 64.0 63.2 0.37

Race/Ethnicity (%)Caucasian 79.1 80.2

0.33Africa-American 11.8 11.3

Hispanic 5.6 5.6

Asian 2.1 1.9

Hypertension (%) 84.6 85.3 0.29

Hypercholesterolemia (%) 82.2 82.2 0.95

Smoking (Current & Former) (%) 61.6 62.7 0.20

Diabetes (%) 39.0 39.7 0.41

Family History of CAD (%) 56.7 57.9 0.17

Heart Failure (%) 8.6 8.8 0.64

Prior MI (%) 42.5 43.3 0.35

Prior PCI (%) 31.9 30.4 <0.05

Prior CABG (%) 13.1 13.5 0.47

Prior Stroke or PVD (%) 17.3 18.4 0.09

Creatinine (mg/dl) 1.15+/-0.88 1.15+/-0.87 0.63

GFR ml/min/1.73 m2 76.5+/-34.1 76.3+/-27.1 0.63

BMI kg/m2 32.7+/-21.9 33.2+/-24.5 0.22

PresentationClinical Characteristics Procedure Status

No-SOS(%)

SOS(%)

Elective 75.9 76.2Urgent 22.7 19.4*Emergency

0.36 0.57*

No-SOS(%)

SOS(%)

STEMI 2.8 3.0NSTEMI 25 26Unstable Angina 37 35Stable Angina 14 14Atypical Chest Pain 5 5Other 17 17

* P < 0.05

No-SOS(%)

SOS(%)

P-Value

Staged 26.1 68.0<0.000

1Lab Visits /Index PCI 1.28 1.73 <0.0001

Single Vessel PCI 80.0 81.9

Multi-Vessel PCI 21.0 22.1

Stent Use     =0.03

DES only 71.9 73.7

BMS only 19.9 19.3

Mixed DES and BMS 4.3 3.4

Balloon only 3.9 3.6

Patient SuccessP=0.007

Lesion SuccessP=0.04

PCI Success: <20% residual stenosis and TIMI 3 flow

Hospitals without

SOSn (%)

Hospitals with

SOSn (%)

Difference in

rate (%)

Asymptoticone-side

95%CI (%)

P Value for Noninferior

ity

N 14,149 4718

Death 132 (0.9 ) 46 (1.0 ) -0.04 -0.31- 0.23 0.004

Hospitals without

SOS

Hospitals with SOS

rate (%)

Asymptotic

one-side95%CI (%)

P Value for Noninferiori

ty

P Value for

Superiority

N 14,149 4718Death 3.2 % 3.2 %

TVR 6.5 % 5.4 % 0.0098

Q wave MI 3.1 % 3.1 %

MACE 12.1 % 11.2 % 0.92 0.04- 1.80 0.05

Hospitals without SOS(N=14,149)

(%)

Hospitals with SOS(N=4718)

(%)

P

All CABG 0.6 % 1.5 % <0.0001 -- Emergency CABG 0.1 % 0.2 %Bleeding 3.4 % 3.2 %Vascular repair 0.4 % 0.4 %Stroke 0.3 % 0.2 %Renal insufficiency 0.5 % 0.4 %Unplanned Catheterization

4.3 % 3.2 %<0.001

ASR 2.7 % 2.7 %

Hospitals without

SOSN=14,149

Hospitals with SOSN=4718

P

All CABG 1.5 % 2.3 % <0.001 Emergency CABG 0.1 % 0.2 %Bleeding 5.3 % 5.2 %Vascular repair 1.1 % 1.2 %Stroke 0.6 % 0.5 %Renal insufficiency 0.9 % 0.8 %Unplanned Catheterization

14.9 % 12.0 % <0.0001

ASR 8.5 % 7.0 % <0.01

Summary Six weeks and 9 months after PCI, the incidence of death, myocardial infarction,

bleeding, stroke, renal failure and vascular repair is similar at hospitals with and without on-site cardiac surgery.

Compared with patients randomized to hospitals with on-site cardiac surgery, in patients allocated to hospitals without on-site cardiac surgery

PCI failure is higher Use of BMS is higher Staged procedures are more frequent A greater number of catheterization laboratory visits is required to complete PCI Frequency of cardiac surgery, including emergency CABG, is higher

Compared with patients randomized to hospitals with on-site cardiac surgery Six week mortality is non-inferior and Nine month MACE

are non-inferior at hospitals without on-site cardiac surgery Incidence of target vessel revascularization is higher at hospitals

without on-site cardiac surgery.

Conclusions

In hospitals without on-site cardiac surgery that complete a formal PCI development program, adhere to C-PORT participation requirements, and whose outcomes are monitored the outcomes of non-primary PCI are non-inferior to outcomes at hospitals with surgery on-site.

Selected Length of Stay Days

Resource Type SOS(n=4569)

No SOS(n=13,704)

P-value

Index Procedures      

Diagnostic catheterization 2.22 0.12 <.0001PCI procedure 1.48 0.19 <.0001PCI + diagnostic cath 0.03 2.83 <.0001 Total Index 3.73 3.14 <.0001Follow-up Period      

Total revascularizations 0.33 0.31 0.74Emergency department 0.26 0.29 0.03 Total Follow-Up 2.48 2.62 0.17Cumulative 9 Months      

Total length of stay 6.20 5.76 0.002

Total Medical Costs

Cost Type SOS No SOS P-value

All Subjects ($)  (n=4569) (n=13,704)  

Index procedure 18,975 19,840 0.12Follow-up period 5015 5620 0.001Cumulative 9 months 23,991 25,450 0.02

Index Procedure Medical Costs:By Cost Center Group

Cost Type SOS No SOS P-Value

All Patients ($)

Non-ICU Room 1091 680 <.0001

ICU Room 2280 3032 <.0001

Total Room 3371 3692 0.001

Cardiac Procedure 8308 9189 0.07

Other costs 7297 6960 0.15

Total Costs 18,975 19,840 0.12

Selected Length of Stay Days: High and Low Volume Sites

Cost Type SOS No SOS P-value

High Volume Sites (n=2389) (n=7161) 

Index procedure 3.87 3.06 <.0001

Follow-up period 2.46 2.42 0.75

Cumulative 9 months 6.33 5.48 <.0001

Low Volume Sites (n=2106)  (n=6324)

Index procedure 3.57 3.24 0.0002

Follow-up period 2.50 2.85 0.02

Cumulative 9 months 6.07 6.09 0.89

High

Low

Total Medical Costs: High and Low Volume Sites

Cost Type SOS No SOS P-value

High Volume Sites (n=2389) (n=7161)   

Index procedure 18,666 19,912 0.18

Follow-up period 5172 5453 0.23

Cumulative 9 months 23,838 25,365 0.15

Low Volume Sites (n=2106)  (n=6324)

Index procedure 19,365 19,807 0.41

Follow-up period 4857 5836 0.0004

Cumulative 9 months 24,222 25,643 0.02

High

Low

C-PORT Results and PolicyC-PORT projects were developed and

implemented to help determine if, within specific parameters, PCI at hospitals without on-site cardiac surgery can be performed safely and effectively.

The purpose was to present to healthcare policymakers another tool to potentially use to in providing optimal access to the best care.

C-PORT Results and PolicyWhat CPORT Says What CPORT Does NOT

sayPCI can be performed

safely and effectively in hospitals without SOS

Hospitals without SOS can safely perform PCI within specific parameters and circumstances

The ability to perform PCI at hospitals without SOS can help policymakers optimize access to PCI service.

PCI should be performed in hospitals without SOS

Any hospital without SOS can safely perform PCI

What the optimal geographic distribution of PCI services should be.

Framing the DebateQuality

Access

Cost

Customized to locality

C-PORT Residual ConcernsC-PORT institutional, practitioner and device inclusion and

exclusion criteria will be forgotten, dismissed and/or ignored.The PCI development program will not be properly

implemented prior to startup of PCI at “new” hospitalsThe human and IT infrastructure supporting data collection

will deteriorate ; quality monitoring and management will degrade

Definition of quality will be compromisedUnintended consequences :

expansion of PCI services will dilute the volume/experience/quality of existing programs

institutional volume criteria will lead to inappropriate procedures