Predictors of in-ICU length of stay among congenital heart defect patients using artificial intelligence model: A pilot study

Objective

This study aims to develop a predictive model using artificial intelligence to estimate the ICU length of stay (LOS) for Congenital Heart Defects (CHD) patients after surgery, improving care planning and resource management.

Design

We analyze clinical data from 2240 CHD surgery patients to create and validate the predictive model. Twenty AI models are developed and evaluated for accuracy and reliability.

Setting

The study is conducted in a Brazilian hospital's Cardiovascular Surgery Department, focusing on transplants and cardiopulmonary surgeries.

Participants

Retrospective analysis is conducted on data from 2240 consecutive CHD patients undergoing surgery.

Interventions

Ninety-three pre and intraoperative variables are used as ICU LOS predictors.

Measurements and main results

Utilizing regression and clustering methodologies for ICU LOS (ICU Length of Stay) estimation, the Light Gradient Boosting Machine, using regression, achieved a Mean Squared Error (MSE) of 15.4, 11.8, and 15.2 days for training, testing, and unseen data. Key predictors included metrics such as "Mechanical Ventilation Duration", "Weight on Surgery Date", and "Vasoactive-Inotropic Score". Meanwhile, the clustering model, Cat Boost Classifier, attained an accuracy of 0.6917 and AUC of 0.8559 with similar key predictors.

Conclusions

Patients with higher ventilation times, vasoactive-inotropic scores, anoxia time, cardiopulmonary bypass time, and lower weight, height, BMI, age, hematocrit, and presurgical oxygen saturation have longer ICU stays, aligning with existing literature.

Initial experience with del Nido cardioplegia solution at a Pediatric and Congenital Cardiac Surgery Program in Brazil

OBJECTIVE

The aim of this study was to evaluate outcome measures between our standard multidose cardioplegia protocol and a del Nido cardioplegia protocol in congenital heart surgery patients.

METHODS

Retrospective single-center study including 250 consecutive patients that received del Nido cardioplegia (DN group) with a mandatory reperfusion period of 30% of cross clamp time and 250 patients that received a modified St. Thomas' solution (ST group). Groups were matched by age, weight, gender, and Risk Adjustment for Congenital Heart Surgery (RACHS-1) scores. Preoperative hematocrit and oxygen saturation were also recorded. Outcomes analyzed were the vasoactive inotropic score (VIS), lactate, ventilation time, ventricular dysfunction with low cardiac output syndrome (LCOS), intensive care unit (ICU) length of stay (LOS), hospital LOS, bypass and aortic cross-clamp times, and in-hospital mortality.

RESULTS

Both groups were comparable demographically. Statistically significant differences (p ⩽ 0.05) were noted for cardiac dysfunction with LCOS, hematocrit at end of surgery (p = 0.0038), VIS on ICU admission and at end of surgery (p = 0.0111), and ICU LOS (p = 0.00118) with patients in the DN group having more desirable values for those parameters. Other outcome measures did not reach statistical significance.

CONCLUSION

In our congenital cardiac surgery population, del Nido cardioplegia strategy was associated with less ventricular dysfunction with LCOS, a lower VIS and decreased ICU LOS compared with patients that received our standard myocardial protection using a modified St. Thomas' solution. Despite the limitation of this study, including its retrospective nature and cohort size, these data supported our transition to incorporate del Nido cardioplegia solution with a mandatory reperfusion period as the preferred myocardial protection method in our program.

Functional lymphatic reserve capacity is depressed in patients with a Fontan circulation

Background

Lymphatic abnormalities play a role in effusions in individuals with a Fontan circulation. Recent results using near‐infrared fluorescence imaging disclosed an increased contraction frequency of lymphatic vessels in Fontan patients compared to healthy controls. It is proposed that the elevated lymphatic pumping seen in the Fontan patients is necessary to maintain habitual interstitial fluid balance. Hyperthermia has previously been used as a tool for lymphatic stress test. By increasing fluid filtration in the capillary bed, the lymphatic workload and contraction frequency are increased accordingly. Using near‐infrared fluorescence imaging, the lymphatic functional reserve capacity in Fontan patients were explored with a lymphatic stress test.

Methods

Fontan patients (n = 33) were compared to a group of 15 healthy individuals of equal age, weight, and gender. The function of the superficial lymphatic vessels in the lower leg during rest and after inducing hyperthermia was investigated, using near‐infrared fluorescence imaging.

Results

Baseline values in the Fontan patients showed a 57% higher contraction frequency compared to the healthy controls (0.4 ± 0.3 min⁻¹ vs. 0.3 ± 0.2 min⁻¹, p = 0.0445). After inducing stress on the lymphatic vessels with hyperthermia the ability to increase contraction frequency was decreased in the Fontan patients compared to the controls (0.6 ± 0.5 min⁻¹ vs. 1.2 ± 0.8 min⁻¹, p = 0.0102).

Conclusions

Fontan patients had a higher lymphatic contraction frequency during normal circumstances. In the Fontan patients, the hyperthermia response is dampened indicating that the functional lymphatic reserve capacity is depressed. This diminished reserve capacity could be part of the explanation as to why some Fontan patients develop late‐onset lymphatic complications.

Aerobic exercise and inspiratory muscle training increase functional capacity in patients with univentricular physiology after Fontan operation: A randomized controlled trial

BACKGROUND

The effect of exercise training and its mechanisms on the functional capacity improvement in Fontan patients (FP) are virtually unknown. This trial evaluated four-month aerobic exercise training and inspiratory muscle training on functional capacity, pulmonary function, and autonomic control in patients after Fontan operation.

METHODS

A randomized controlled clinical trial with 42 FP aged 12 to 30 years and, at least, five years of Fontan completion. Twenty-seven were referred to a four-months supervised and personalized aerobic exercise training (AET) or an inspiratory muscle training (IMT). A group of non-exercise (NET) was used as control. The effects of the exercise training in peak VO₂; pulmonary volumes and capacities, maximal inspiratory pressure (MIP); muscle sympathetic nerve activity (MSNA); forearm blood flow (FBF); handgrip strength and cross-sectional area of the thigh were analyzed.

RESULTS

The AET decreased MSNA (p = 0.042), increased FBF (p = 0.012) and handgrip strength (p = 0.017). No significant changes in autonomic control were found in IMT and NET groups. Both AET and IMT increased peak VO2, but the increase was higher in the AET group compared to IMT (23% vs. 9%). No difference was found in the NET group. IMT group showed a 58% increase in MIP (p = 0.008) in forced vital capacity (p = 0.011) and forced expiratory volume in the first second (p = 0.011). No difference in pulmonary function was found in the AET group.

CONCLUSIONS

Both aerobic exercise and inspiratory muscle training improved functional capacity. The AET group developed autonomic control, and handgrip strength, and the IMT increased inspiratory muscle strength and spirometry.

CLINICAL TRIAL REGISTRATION

ClinicalTrials.gov Identifier: NCT02283255.