Bibliographic record
Abstract
This editorial refers to ‘CALL-K score: predicting the need for renal replacement therapy in cardiogenic shock’ by E. Rodenas-Alesina et al., https://doi.org/10.1093/ehjacc/zuac024 Despite more than 40 years of investigation, there remains only one proven therapy to improve outcomes in cardiogenic shock (CS), which is urgent coronary revascularization of the culprit lesion in patients with acute myocardial infarction (AMI).1,2 However, no treatment has yet shown to improve outcomes in CS not caused by AMI, including percutaneous mechanical circulatory support devices, which have yet to prove their efficacy in CS irrespective of aetiology. Randomized controlled trials have proven to be exceedingly challenging to conduct in the setting of CS, and therefore, registries have been increasingly useful as we search for actionable insights into management for these patients. Rodenas-Alesina et al.3 from the Peter Munk Cardiac Centre in Toronto, Canada, offer new guidance from a large single quaternary care centre with a prospective registry of CS patients collected from 2014 to 2020 and encompassing 1030 patients. Approximately half of the patients had a diagnosis of chronic heart failure prior to admission, and nearly 21% had CS in association with AMI. The authors focused on the prediction of renal replacement therapy (RRT) and its association with mortality over the short and longer term. They derived a novel score (CALL-K) which included terms for the complete blood count, age, elevated lactate, use of loop diuretics, and admission glomerular filtration rate. The resultant score is simple to calculate and is marginally better than simple serum creatinine alone to predict the need for RRT.3 RRT (in any form) was associated with major morbidity and mortality including longer length of stay, rate of sepsis and pneumonia, and lower likelihood of receiving definitive therapies such as heart transplant or ventricular assist device. Moreover, the hazard extended to 1-year follow-up, and it is clear that RRT during the index hospitalization identified a subgroup with extremely high 1-year mortality with only 30.9% of such patients surviving a year post hospitalization and nearly 72% of survivors on chronic hemodialysis following discharge. Where does the CALL-K score fit in the toolbelt of the clinician managing CS patients? It is tempting to think that the score will be utilized to give predictions to the team and patient’s family, but this is likely not to be the case. Other scores such as the CARDSHOCK,4,5 IABP- SHOCK-2,6 CLIP,7 and CSS8 risk scores are more extensively validated and yet to date may remain of more academic than practical utility. However, there are several lessons to be gleaned from the CALL-K score and analysis (see Figure 1). With non-ischaemic shock being the predominant form of CS in this large single-centre registry, like others,9,10 it questions the current practice of randomized CS trials to exclusively enrol AMI patients. Although AMI-CS seems to be a more homogenous population and enrolling other shock entities might be more challenging, expanding the evidence base to non-ischaemic CS is an issue which has to be tackled. While there is much debate over the utility of pulmonary artery catheters for the assessment and ongoing management of CS patients,11–13 it is notable that only 34.3% of patients underwent such monitoring in a centre with full access to advanced therapies including heart transplantation. It is possible that the patients’ volume status was evident without invasive monitoring, but the majority of the patients were in Society for Cardiac Angiography and Intervention (SCAI) State D shock,14 with a third receiving mechanical ventilation and more than 20% with IABP or extracorporeal life support ongoing. The authors note that 57 patients undergoing RRT (total 123 patients, 46.3%) also had a PA catheter, which allowed further insights to be obtained. RRT patients with invasive hemodynamic monitoring had significantly higher central venous pressure and higher CVP/pulmonary capillary wedge pressure ratio and a lower pulmonary artery pressure index. In aggregate, these findings suggest that right ventricular failure and congestion are key mediators of mortality in CS and that measurement of invasive haemodynamics may assist in phenotyping patients. Others have noted similar findings.13,15 All of the components of the CALL-K risk score are readily available at the time of admission. Other risk scores such as SCAI stage can also be assessed in an algorithmic fashion from vital signs and information about therapies employed.16,17 Electronic medical records could be enhanced to provide real-time feedback about patient status which would allow the clinician to integrate the information into the overall management of the patient. It might also trigger warnings about a patient at risk of deteriorating and could therefore be used to allocate resources towards patients at higher risk. It is likely that in some cases, the panel of risk scores will provide value, whereas in other cases, the clinical picture will not agree with the computerized evaluation. Nevertheless, particularly as the morbidity and cost associated with extensive critical care treatments is significant, it would be ideal to provide the treatment team with easily obtainable insights to allow prudent decisions to be made prospectively rather than suffer retrospective regret. Central illustration: lessons and opportunities arising from the CALL-K score. In summary, the current report adds to the growing body of evidence from prospective registries of CS patients and also highlights areas with unmet needs. Clearly, we have much to learn, and the lack of positive progress in improving the mortality of such patients should not serve to frustrate us but rather to galvanize and energize clinicians and researchers to push on and find solutions to these vexing problems. Future registries could consider prospective assignment of SCAI shock stage, as well as serial data on outcomes associated with these patients along the continuum of care. At least one such registry has begun,18 and the opportunity for others to emulate this approach beckons. We have weapons and ammunition in the war against CS; the time is now to call for backup in the form of registries which will hopefully illuminate the path forward. Conflict of interest: D.A.B. has consulted for Livanova, Getinge, Abbott and Abiomed. He is on the steering committee for Procyrion and CareDx. B.S. reports no relevant conflicts.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.003 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".