Related Experiment Video
Updated: Aug 9, 2026
07:28
Real-time Pressure-volume Analysis of Acute Myocardial Infarction in Mice
Published on: July 2, 2018
Missing Lactate Values in Critical Care Registries: A Multiple Imputation Framework for Cardiogenic Shock Research
Robert Thiesmeier1, Jeong-Gun Park2, Siddharth M Patel2
1TIMI Study Group, Division of Cardiovascular Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, USA; Department of Global Public Health, Karolinska Institutet, Stockholm, Sweden.
Journal of Clinical Epidemiology
|July 22, 2026
Summary
Missing lactate measurements in critical care registries introduce bias. A structured multiple imputation framework improves risk prediction accuracy for patients with cardiogenic shock, enhancing real-world evidence reliability.
Area of Science:
- Cardiovascular epidemiology
- Critical care medicine
- Biomarker analysis
Background:
- Lactate is a key prognostic biomarker in cardiac intensive care and risk models.
- Missing lactate values are common in registries due to logistical and clinical factors.
- Current methods for handling missing lactate (exclusion, indicator variables) can introduce bias.
Purpose of the Study:
- To develop and evaluate a structured methodological framework for addressing missing lactate values in clinical research.
- To characterize the mechanisms of lactate missingness in a large cardiac intensive care unit registry.
- To assess the impact of imputing missing lactate on risk reclassification in prognostic models.
Main Methods:
- Utilized data from 17,993 admissions (2017-2021) from the Critical Care Cardiology Trials Network registry.
- Characterized lactate missingness and developed a multiple imputation model.
- Evaluated risk reclassification using the IABP-SHOCK II risk score with imputed lactate values.
Main Results:
- Lactate missingness was associated with patient severity, indicating data were not missing completely at random.
- The multiple imputation model incorporated identified predictors of missingness.
- Imputing lactate values led to significant risk reclassification in the IABP-SHOCK II score, highlighting bias in naive methods.
Conclusions:
- Missing lactate measurements contribute to selection and measurement bias in critical care registries.
- A structured imputation framework effectively mitigates bias.
- Improved reliability of real-world evidence in cardiovascular epidemiology is achieved through principled handling of missing biomarker data.
