Most STEMI Patients Head to ICU After PCI, but Not All Need to Go

The gap between who’s sent and who needs to be there suggests a need for a risk score to triage higher-risk patients to the ICU.

Most STEMI Patients Head to ICU After PCI, but Not All Need to Go


(UPDATED) Despite near universal admission of STEMI patients to the intensive care unit (ICU) following PCI, a new study suggests there is a “discrepancy” between who is sent to the unit and who actually needs to be there.

In the analysis, researchers showed that while more than 80% of patients who underwent uncomplicated primary PCI for STEMI are admitted to the ICU after the procedure, just 16% developed complications that would justify the admission to the cardiac intensive care unit (CCU).

“It’s hard for us to say that we’re ‘needlessly’ admitting patients to the ICU,” lead investigator Jay Shavadia, MD (Duke Clinical Research Institute, Durham, NC), told TCTMD. “I don’t think that’s a correct statement. I think 16% is high, and what we’re trying to highlight is that it’s true we’ve identified a large discrepancy in our current practice, which is almost universal admission versus who actually needs to be there.”

Given this gap, Shavadia said there is a need for “risk-based triage,” an approach that would focus on factors such as the delay in reperfusion in order to optimize ICU use for patients with STEMI. A stratification tool would allow higher-risk patients to be sent to the resource-intensive ICU and allow those at lower risk for post-PCI complications to be monitored elsewhere, said Shavadia.

“We need a better way to figure out who needs to be [in the ICU],” he said.

Rapid Reperfusion and Risk of Complications

The new study, published April 15, 2019, in JACC: Cardiovascular Interventions, is an analysis of 19,507 patients aged 65 years and older hospitalized with STEMI at 707 hospitals in the National Cardiovascular Data Registry (NCDR) Chest Pain-MI Registry. Of these patients, all of whom were hemodynamically stable, 82% were sent to the ICU following uncomplicated primary PCI during the index admission. Median length of stay in the ICU was 1 day.

Despite the large proportion of patients sent to the ICU, just 16% of the total population developed complications after PCI that would require an ICU stay. Of the 16,047 patients admitted to the ICU, 4.1% died, 4.1% developed cardiac arrest, 9.8% developed cardiogenic shock, and 1.0% had a stroke. Additionally, 4.6% developed atrioventricular block requiring treatment and 6.4% developed respiratory failure. For the 3,460 patients not transferred to the ICU, 7.8% developed complications that eventually required treatment in the ICU. 

To TCTMD, Shavadia said that despite the gap between those sent to intensive care and the complication rate, the number of complications that did develop is not trivial given the patient population.

“These are people who didn’t have shock, who didn’t have cardiac arrest at admission, who really didn’t have any procedural complications, and who were all pretty quickly treated,” he said. “For 16% to develop a complication when they were all initially uncomplicated, it tells us there are a group of people who do develop complications despite being stable when they first got into the hospital.”    

Overall, the median time from first medical contact (FMC)-to-device time was 79 minutes, with 4,298 patients treated within 60 minutes or less. Patients with longer FMC-to-device times had a higher rate of developing ICU-requiring complications (P < 0.001 for trend). Those with longer reperfusion times were at a particularly higher risk of dying in-hospital, or developing cardiac arrest, shock, or respiratory failure. In a multivariate-adjusted analysis, an FMC-to-device time of 61-90 minutes was associated with a significant 13% higher risk of ICU-level complications compared with an FMC-to-device time of 60 minutes or less. Those with reperfusion times exceeding 90 minutes had a significant 22% higher risk of ICU-requiring complications.

“The first medical contact-to-device time is just one variable that we used,” said Shavadia. “We thought that ischemic time is probably the most important determinant of STEMI outcomes . . . but there are certainly more variables than ischemic time.”  

Sean van Diepen, MD (University of Alberta, Canada), who was not involved in the present study but has investigated ICU utilization in NSTE-ACS patients, said that “rapid reperfusion does lower the risk of STEMI complications” and that the next step will be to derive and validate a model for predicting the risk of in-hospital complications in hemodynamically stable STEMI patients who underwent PCI.

In addition to FMC-to-device time, other factors that may contribute to the development of complications include symptom onset to presentation, past medical history, anterior versus nonanterior MI, post-PCI TIMI flow, electrocardiographic reperfusion, left ventricular ejection fraction, and degree of clinical heart failure, van Diepen told TCTMD.

“Reducing low-risk STEMI admission to the cardiac ICU can potentially reduce health care costs and lower critical care capacity strain,” he said. “There is a growing body of evidence to suggest that cardiac ICUs are being over-used for low-acuity patients. I think it is reasonable for hemodynamically stable patients with low-risk STEMI to be admitted to a non-cardiac ICU environment after primary PCI provided they are in a telemetry environment with trained nursing and allied health staff with institutional protocols and pathways to escalate care to address cardio-respiratory complications.”

Adnan Chhatriwalla, MD (Saint Luke’s Mid America Heart Institute, Kansas City, MO), who was also not involved in the study, agreed that rapid reperfusion lowers the risk for ICU-requiring complications. Other factors include the stability of the patient leaving the cath lab, he told TCTMD. For example, the presence of hypoxia, pulmonary edema, elevated left ventricular pressures, bleeding, hypotension, and arrhythmia are all considerations for sending the patient to ICU rather than a step-down unit or telemetry-enabled wards. Age and infarct vessel are also important considerations.

“It’s standard for us, although we have noticed in the NCDR CathPCI Registry that other sites must not routinely be admitting patients to the ICU based on length-of-stay data,” said Chhatriwalla in an email. “I tell patients/family that the reasons for ICU, in an otherwise stable patient, are because of risks for vessel reocclusion, congestive heart failure, hypotension, or arrhythmias.”    

Standard Practice

Admission to the cardiac ICU for monitoring is standard practice in most hospitals, said Shavadia. In fact, the European Society of Cardiology (ESC) guidelines for the treatment of STEMI, which were published in 2017, recommend 24 hours in the ICU. The coronary/cardiac care unit was designed to reduce the risk of ventricular fibrillation/ventricular tachycardia (VF/VT) associated with acute MI, but there have been significant changes in the treatment of STEMI patients, chief among them faster reperfusion that lessens ischemic time and cuts the risk of post-PCI complications, said Shavadia. 

To TCTMD, van Diepen said the ESC recommendation is largely based on expert opinion and is a reflection of “historical practices that have been slow to adapt to improved STEMI revascularization and systems of care that have reduced mortality and in-hospital complications.” In the NCDR analysis, there was extensive inter-hospital variability in the percentage of primary PCI STEMI patients sent to the ICU, ranging from 5% to 100%, which suggests some hospitals are adopting. He added that that many centers might also lack stepdown units or telemetry-enabled ward beds to appropriately monitor low-risk STEMI patients.

In an editorial, Suartcha Prueksaritanond, MD, and Ahmed Abdel-Latif, MD, PhD (both from University of Kentucky, Lexington), point out that the contemporary cardiac care unit is often a destination for complex patients with advanced cardiovascular and comorbid conditions, and the need for such critical care in the ICU is increasing. For this reason, hospitals need to prioritize their limited resources by adopting risk-stratification tools to determine who should and shouldn’t go to the ICU, write the editorialists. 

“The high ICU utilization pattern, despite declining complications following primary PCI, calls for a new approach,” state Prueksaritanond and Abdel-Latif. “This is particularly important because the overall health care cost continues to grow and calls for optimal resource utilization to prevail.”

  • The authors and editorialists report no conflicts of interest.

We Recommend