The CT FIRST Trial: Should We Pan-CT After ROSC?

Background: Achieving ROSC in out of hospital cardiac arrest (OHCA) is no easy feat but, care doesn’t end with ROSC. Post-ROSC management is nuanced and challenging but helps to ensure good outcomes. Identification of the underlying cause of the cardiac arrest is a critical area of focus in post-arrest care. Although myocardial infarction, dysrhythmias and pulmonary emboli are common pathologies to consider, there are a host of other causes including subarachnoid hemorrhage, trauma and electrolyte disturbances.

In theory, rapid identification of the underlying cause should improve outcomes by allowing clinicians to tailor management. Advanced imaging post-arrest is a possible modality to achieve this end.

Article: Branch KHR et al. Diagnostic yield, safety, and outcomes of Head-to-pelvis sudden death CT imaging in post arrest care: The CT FIRST cohort study. Resus 2023. PMID: 37019352

Clinical Question: Does protocolized head-to-pelvis sudden death CT (SDCT) scanning improved the diagnostic yield post-ROSC in comparison to standard practice?

Population: Patients successfully resuscitated from OHCA without a clear diagnosis on presentation who were stable for CT.

Outcomes:

  • Primary: Diagnostic yield (defined as the number of patients with an adjudicated diagnosis that was the presumed cause for OHCA) of SDCT in comparison to standard practice in identifying the cause of cardiac arrest.
  • Secondary:
    • Time to adjudicated diagnosis.
    • Percentage of correct diagnoses
    • Diagnosis of time critical diagnoses
    • Delayed diagnosis > 6 hours from hospital arrival
    • Safety: Acute Kidney Injury

Intervention: “The SDCT scan protocol consisted of three CT scans: 1) a non-contrast head CT, 2) retrospective ECG-gated thoracic CT contrast angiogram for most of the cardiac cycle (initially 20–90% and later 30–80% of the cardiac cycle to reduce radiation dose), and 3) a venous phase, non-ECG gated, spiral abdominal and pelvis CT,” performed within 6 hours of presentation.

Control: Standard approach to diagnosis post-arrest but would be eligible for SDCT.

Design: Prospective, observational, before and after implementation of a protocol study.

Excluded:

  • Obvious cause for OHCA prior to SDCT or on hospital arrival.
  • Indication for emergency invasive coronary angiography or had coronary angiography within 1 hour of arrival.
  • Known obstructive coronary artery disease or known coronary stent.
  • Known cardiac defibrillator.
  • Pre-existing DNR order.
  • Severe renal dysfunction.

Primary Results

    • Standard care enrollment period
      • 143 of 273 patients with ROSC after OHCA were eligible for enrollment.
      • Most common reason for exclusion: Would not have been able to undergo SDCT.
    • SDCT enrollment period
      • 111 of 307 patients with ROSC after OHCA were eligible for enrollment. Several were later excluded.
      • Most common reason for exclusion: Patient was not scanned.
    • CT scanning was common in the standard care cohort (84% received at least one CT – usually NCHCT)

Critical Findings:

Standard Care Group

SDCT Group

Diagnosis (Primary Outcome)

75% (107/143)

92% (96/104)

Statistically Significant

Time to Diagnosis

14.1 (IQR 2.2 – 69.5)

3.1 (IQR 1.4-12.9)

Statistically Significant

Identified Time-Critical Diagnosis

24% (34/143)

32% (33/104)

Not Statistically Significant

Delayed Ascertainment of Time Critical Diagnosis

62% (21/34)

12% (4/33)

Statistically Significant

Survival to Hospital Discharge

44% (63/143)

42% (44/104)

Not Statistically Significant

Acute Kidney Injury

24% (34/143)

26% (27/104)

Not Statistically Significant

Strengths:

  • Investigates an important area with limited prior research.
  • Most baseline demographics were similar between the two groups.
  • Post-arrest care (aside from the studied intervention) appears to be similar between the two groups.

Limitations:

  • Observational, before and after studies cannot show causality; they can only reveal associations.
  • Small sample size of patients.
  • Risk of selection bias based on the large number of patients that were not included.
  • The authors did not give detailed information on why patients were excluded particularly in the SDCT enrollment period when the vast majority of ROSC patients were not included.
  • No 95% confidence intervals are supplied for the primary or secondary outcomes making it impossible to assess confidence of the findings.

Discussion

  • Framing this as a diagnostic study reveals that SDCT was successful in what it is intended to do: identify pathology. However, it’s important to determine the benefit of the improvement in diagnosis.
    • An improvement in diagnostic accuracy is important if it leads to improved outcomes for the patient. In the absence of improved outcomes, the costs (however modest) likely outweigh the benefits.
    • Costs include actual cost of the scans as well as delaying other patients’ scans but also includes time taken away from resuscitation to move the patient to and from CT. The cost difference may be modest here as many of the patients in the standard care cohort were getting CTs as well.
  • In real life, many of the exclusion criteria will not be known at the point of first contact (ie DNR order, history of obstructive coronary artery disease).
  • Treatment implications of the study
    • Before and after studies (in terms of treatment) can be biased by other advances in care that occur during the intervening time. For example, if some other proven intervention became available to the after group, this group would have more favorable outcomes regardless of the intervention being studied.
    • Bystander CPR, a known predictor of good outcomes, was more common in the SDCT cohort than in the standard care cohort.
  • There is no discussion of how many of the diagnoses made by SDCT could have been made clinically instead.
    • This is similar to what we see in the trauma “pan-scan” literature: if you don’t look for clinical pathology, CT scan will identify a number of things you didn’t recognize.
  • Many patients in the standard care cohort underwent CT imaging. This may reduce the impact of protocolized SDCT in comparison to an institution where CT use post-ROSC is minimal.
  • Majority of patients excluded in the SDCT cohort were excluded for inability to get to CT within 6 hours. A protocol encouraging CT within 6 hours may increase risk for the patient by prioritizing imaging over resuscitatin.

Authors Conclusions: “SDCT scanning early after OHCA resuscitation safely improved the efficiency and diagnostic yield for causes of arrest compared to the standard of care alone.”

Our Conclusions: The authors conclusions overstep what their data can reveal. SDCT was associated with a higher and more rapid diagnostic yield post-arrest than standard practice. However, this improvement did not translate to improved patient outcomes.

Potential to Impact Current Practice: This data should not change clinical practice. High-quality, randomized studies are needed to further elucidate the utility of SDCT.

Bottom Line: Routine SDCT should not be part of standard practice in patients with ROSC after OHCA. Clinicians should use their clinical assessment to guide advanced imaging in this group.

References:

  1. Branch KHR et al. Diagnostic yield, safety, and outcomes of Head-to-pelvis sudden death CT imaging in post arrest care: The CT FIRST cohort study. Resus 2023. PMID: 37019352

Post Peer Reviewed By: Salim R. Rezaie, MD (Twitter: @srrezaie)

Cite this article as: Anand Swaminathan, "The CT FIRST Trial: Should We Pan-CT After ROSC?", REBEL EM blog, June 1, 2023. Available at: https://rebelem.com/the-ct-first-trial-should-we-pan-ct-after-rosc/.

Like this article?

Share on Facebook
Share on Twitter
Share on Linkdin
Share via Email

Want to support rebelem?

Sponsored