Introduction
Hemophagocytic syndrome (HPS) is a potentially fatal systemic inflammatory disorder that arises from the relentless activation of cytotoxic T cells, natural killer cells, and the excessive stimulation of macrophages. The incidence of HPS has notably surged by an alarming 11% annually in England from 2003 to 2018, reflecting a striking fourfold increase over this 16-year period. This alarming trend includes a 14% annual increase in individuals aged 15 to 54 years and a very concerning 16% rise among those aged 55 and older. In adults, HPS frequently occurs secondary to various conditions including infectious, malignant, and autoimmune diseases. The clinical and laboratory findings related to HPS are largely nonspecific, typically featuring symptoms like fever, hepatosplenomegaly, cytopenia, and hyperferritinemia. A significant number of patients are admitted to intensive care units (ICUs), necessitated by serious organ dysfunction. Mortality rates in patients with HPS who progress to multiorgan dysfunction are exceedingly high, ranging from 40% to 80%, indicating the severity of this condition.
Life-threatening clinical manifestations that have been identified in HPS patients include acute liver failure, disseminated intravascular coagulation, acute respiratory failure, decreased levels of consciousness, and seizures. Recent studies have highlighted cardiovascular complications among HPS patients, which include acute heart failure, myocardial infarction, and circulatory failure, with associated mortality rates between 25.6% and 53%. The heart serves as a crucial target of systemic inflammation, revealing that inflammation plays a significant role in the onset of acute heart failure. Various pro-inflammatory cytokines have been implicated in cardiac injury, including TNF-α, IFN-γ, IL-1β, IL-6, IL-17, and IL-18. Notably, elevated levels of TNF-α are linked to detrimental inotropic effects and increased mortality. Additionally, IL-1β has been recognized for its cardiodepressive effects by inducing the release of inducible nitric oxide. In persistent high-grade systemic inflammation, cardiac damage associated with circulatory failure appears almost inevitable, although early immunosuppressive treatment has shown promise in effective management.
Post-COVID-19 pandemic, there is heightened awareness regarding the interplay between cardiac health and hyperinflammation, especially in light of HPS being characterized as a multisystem hyperinflammatory disorder. This rarity in adults means that there is a significant lack of data available to forecast cardiac complications throughout the course of HPS. In this retrospective study, our objective was to discern the factors that predict the development of acute heart failure in ICU-admitted patients with HPS.
Materials and Methods
This retrospective observational study focused on adult critically ill patients diagnosed with HPS according to the HLH-2004 diagnostic criteria in the ICU of Ege University Faculty of Medicine from 2012 to 2023. De novo acute HF was identified through a meticulous review of electronic medical records, necessitating the absence of previous diagnoses of heart failure and associated medication use, alongside the emergence of new and rapidly worsening symptoms per established guidelines. Patients with an existing history of heart failure or those with missing data were excluded from the analysis. The study received approval from the Institutional Ethical Review Board of Ege University Hospital, and we adhered to good clinical practice guidelines and the principles outlined in the Declaration of Helsinki throughout the study. Informed consent was obtained from either the patients or their relatives. Based on diagnostic assessments, HPS etiological diagnoses were determined by a multidisciplinary team comprising an intensivist, rheumatologist, and hematologist. Essential demographic, clinical, and laboratory data were retrieved from electronic medical records, and the cardiothoracic ratio (CTR) was quantified on a PA chest x-ray. This ratio represents the maximal transverse dimension of the heart relative to the maximal transverse thoracic dimension, expressed as a percentage. The primary outcome of our study centered around identifying predictive risk factors for the development of heart failure in HPS patients, while the secondary outcome assessed ICU mortality.
Statistical Analysis
Data from the study were summarized using descriptive statistics, with continuous variables expressed as mean ± standard deviation or median with ranges, based on their distribution. Categorical variables were summarized as counts and percentages. The normality of numerical variables was evaluated using suitable tests and visualizations dependent on sample size and data characteristics. For smaller sample sizes, the Shapiro–Wilk test was utilized, while the Kolmogorov–Smirnov test and Anderson-Darling test were employed for larger sample sizes, respectively. Visual aids such as histograms and Q-Q plots supplemented these determinations of normality. To compare categorical variables, the Pearson chi-square test was used for 2×2 tables with adequate expected cell counts. In scenarios where expected counts fell below 5, Fisher’s exact test was preferred. For RxC tables with expected counts less than 5, the Fisher-Freeman-Halton test was the method of choice.
For comparison between two independent groups, an independent samples t-test was adopted if the numerical variables exhibited a normal distribution; otherwise, the Mann–Whitney U-test was chosen. We conducted a classification and regression tree (CART) analysis to identify predictive models for acute heart failure status among a dataset of 146 participants, validated through 10-fold cross-validation. Model 1 incorporated variables such as age, CTR, etiology, NT-proBNP, CRP, triglyceride, LDH, and ferritin levels recorded at diagnosis. Model 2 included CTR, CRP, albumin, etiology, gender, and LDH. Equal prior probabilities were assigned to each class for the analysis purposes. The Gini criterion was employed for node splitting, with optimal trees selected based on minimum misclassification cost within one standard error.
The rationale for the selection of CART analysis over logistic regression rests upon the assumption of nonlinear relationships and intricate interactions within the dataset. Notably, CART allows for the identification and interpretation of these complex relationships, while its nonparametric nature fosters enhanced flexibility in modeling. A significant advantage of CART is the visual interpretability of results, facilitating the achievement of research objectives. Statistical procedures were performed using Jamovi, JASP, and Minitab software packages, with a statistical significance level set at p ≤ 0.05.
Results
Among the 146 patients studied, 49.3% developed manifestations and symptoms indicative of acute heart failure. A noteworthy observation was the higher female-to-male ratio among patients exhibiting heart failure, with this discrepancy proving statistically significant (59.7% female vs. 40.3% male, p = 0.031). Furthermore, the CTR was significantly elevated in patients with acute heart failure (p = 0.007), establishing a critical correlation. Rheumatological causes predominated among those developing acute heart failure, whereas infectious etiologies were more common in patients without heart failure. Key factors including age, length of hospital stay, time from diagnosis to treatment, time from diagnosis to death, and mortality rates did not exhibit significant disparities between the two patient groups (p > 0.05 for each).
The levels of NT-proBNP, triglycerides, direct bilirubin, and CRP were markedly elevated in patients with acute heart failure.
Model 1: CART Results
The CART analysis identified CTR and age as the primary predictors, resulting in a decision tree with four terminal nodes. This analysis was predicated on CTR ≤ 52.5 and age ≤ 60 years. The further splits distinguished patients based on unique etiologies. In patients with a CTR below 52.5, the probability of not experiencing heart failure was as high as 75.4%. Conversely, in those with a CTR exceeding 52.5, the probability of acute heart failure reached 69.1%, indicating the pivotal role of age. In fact, this risk heightened to 84.3% for patients with a CTR above 52.5 who were also aged 60 years or under. Patients with a CTR above this threshold, aged over 60, exhibited variations according to etiology; those with infection, malignancy, or other causes had a 69.6% chance of not having heart failure, while those with rheumatological causes had an alarming 85.7% chance of experiencing acute heart failure.
Model Performance
The confusion matrix utilized for evaluating model performance indicated that, in the training data, 49 out of 72 acute heart failure cases were correctly predicted, yielding an accuracy of 68.1%. In contrast, 65 out of 74 cases without heart failure received correct predictions, resulting in an accuracy of 87.8%. The overall accuracy across the training dataset was 78.1%. In the test data, the model successfully predicted 43 out of 72 cases of acute heart failure (59.7% accuracy), and 59 out of 74 cases without heart failure were correctly identified, yielding an overall accuracy of 69.9%. Model sensitivity trends showed 68.1% for training data and 59.7% for test data while detailing the accompanying false positive and negative rates.
These findings suggest that while the model demonstrated reasonable correct prediction rates, certain limitations in generalizability should be considered.
Variable Importance
According to our analysis, the relative importance of predictors was quantified, with the CTR rated at 100% as the most significant predictor, followed by age, NT-proBNP, and etiology, thus highlighting the need for ongoing focus on these factors in clinical practice.
Model 2: CART Results
In exploring additional predictive models, CTR and CRP emerged as critical predictors. The analysis produced a decision tree with five terminal nodes, emphasizing the importance of both CTR and CRP levels in assessing heart failure risk.
Model Performance
The model performance evaluation reiterated the need for careful interpretation and suggested that additional factors might be relevant for accurate prediction, warranting further investigation.
Discussion
This study provides a deeper understanding of the predictors of acute heart failure in patients with HPS utilizing CART analysis. The identification of CTR as the paramount predictor signifies its vital role in patient assessment and management. Our findings advocate for the necessity of consistent monitoring of cardiothoracic ratios in patients exhibiting symptoms of HPS, particularly since age and etiology also significantly influence acute heart failure risk.
The implications of such predictors could guide clinical interventions and improve patient outcomes through early identification and tailored treatment strategies.
Conclusion
Our findings emphasize a multifaceted approach to managing patients with HPS, advocating for early interventions driven by risk factors. Understanding predictors like cardiothoracic ratio and inflammatory markers could bedeck future patient management strategies.
Abbreviations
HPS, hemophagocytic syndrome; HF, heart failure; CART, classification and regression tree; CTR, cardiothoracic ratio; NT-proBNP, N-terminal pro-brain natriuretic peptide; CRP, C-reactive protein; LDH, lactate dehydrogenase; ICU, intensive care unit; AUC, area under curve.
Data Sharing Statement
The datasets utilized during the study can be accessed from the corresponding author upon reasonable request.
Ethics Approval and Informed Consent
All participants or their relatives provided informed consent prior to study inclusion. This research received ethical approval from the Institutional Ethical Review Board of Ege University Hospital. The study complied with the standards set by good clinical practice guidelines.
Acknowledgments
We extend our gratitude to contributors for their support in statistical analysis.
Funding
There is no funding to report.
Disclosure
The authors declare no relevant financial or non-financial interests.
Absolutely, let’s dive into this rather intricate world of Hemophagocytic Syndrome (HPS) and its unfortunate sidekick—acute heart failure. A topic that sounds like it was concocted in a medical laboratory after binge-watching medical dramas!
### Introduction
So, what is this “hemophagocytic syndrome”? Sounds like something you’d find in a Harry Potter episode, doesn’t it? It’s actually a potentially fatal inflammatory disorder (the kind that wouldn’t make it to the dinner party invite list). Essentially, it’s when the body’s own immune cells just lose their cool—and that’s putting it mildly. Cytotoxic T cells and natural killer cells are on hyperdrive while macrophages are ready to stage a full-on coup.
And boy, has its popularity been soaring! The incidence in England jumped 11% each year from 2003 to 2018, which means if HPS were a pop star, it would be headlining Glastonbury right about now. We’re talking a staggering rise in cases especially among those in their 30s to 50s, and even the silver foxes over 55 are getting a piece of the action! What’s worse? If HPS decides to throw a party in the ICU (which is not a fun place, mind you), the mortality rates can hit somewhere between a shocking 40% to 80%. And the symptoms? Well, they throw in the works—fever, enlarged organs, and a delightful dash of hyperferritinemia. Sounds like the worst buffet ever!
### Heart Failure and the Dirty Work of Inflammation
Now, let’s talk about what happens when that rampant inflammation decides to crash the cardiac bash—enter acute heart failure (HF). For those who don’t have a PhD in medical jargon, it’s where the heart decides, “Nope, not today!” and stops pumping efficiently. The culprits? Pro-inflammatory cytokines. Think of them as that overly enthusiastic guest who brings too much party poppers and sets the whole place on fire.
Among these the biggest troublemaker is Tumor Necrosis Factor Alpha (TNF-α)—not to be confused with a hipster band name! Elevated TNF-α levels correlate to heart dilation and, surprise surprise, increased mortality! It’s a party of bad news!
What our authors are getting at here is that amidst these chaotic health scenarios, there’s a scary gap in our understanding of how heart failures develop in patients with HPS—like knowing your car makes a funny sound but having no clue what’s wrong underneath the hood.
### Study Overview
Now hold onto your stethoscopes because here comes the study—a retrospective look at HPS patients in ICU, and what the researchers are trying to do is find out what risk factors are associated with acute heart failure. It involves a bunch of statistical wizardry worthy of Hogwarts. They collected data from 146 patients, those brave souls, and sifted through their medical records like detectives on a mission to crack the case of the mysterious, vanishing heart function.
They came up with two models using CART analysis (let’s just say it’s like a decision tree that even a five-year-old could understand). They evaluated a list of suspects: age, cardiothoracic ratio (CTR), various blood markers, the etiology (the cause of HPS), and others. These terms will drive any non-medical person up the wall faster than you can say “emergency room.”
### The Results: Plot Twists Ahead
So, what did they find? A whopping 49.3% of the patients had developed acute heart failure! The female-to-male ratio? Ladies, you dominate—59.7% of those in heart failure were women. And those with rheumatological causes? They were practically handing out invitations!
The analysis revealed CTR and age were the top predictors—if your heart takes up too much space in your chest, beware! It’s as if one’s heart is saying, “Make room, I’m coming through, and I might just take you down with me.” Those with a CTR above a certain value had a high probability of heart failure—remember this next time someone tells you they have a “big heart,” it might not be the compliment they think it is!
### So, What Do We Make of All This?
Let’s face it, understanding HPS and heart failure feels like trying to decode Shakespeare whilst standing on one leg! But why does this matter? For the sake of your health, if the medical world can predict and identify at-risk individuals, it can act swiftly—like a doctor sniffing out a fake cough at a health fair—and intervene before things go south.
With these findings under our capes, there’s a clarion call for further research! We’d need larger studies, possibly even multi-center ones, to make sure our gallant conclusions don’t just land us in the wrong castle. There’s also the ever-popping concern of confounding variables—like trying to watch a Netflix series while someone is playing the trombone on the same couch!
### Conclusion
Here’s the hair-raising takeaway: if you or someone you know has HPS, consider keeping an eye on heart health. The more we arm ourselves with knowledge—think of it as the magical sword of medical wisdom—the better we can slay the dragon of acute heart failure that lurks in the shadows.
In essence, HPS may sound obscure and tricky, but it is a beast we can face head on with the right insights. So stay informed and, as always, be cautious—after all, having a heart is important, but having a healthy one is what we’re really after.
—
And there you have it, folks! Could heart failure be the next big hit on the medical charts? Maybe not, but with each step in research, we get a little closer to understanding this puzzling condition. Stay tuned as we pull back the curtain on future studies, hopefully bringing joy and not just the heartbreak of acute heart failure! Cheers!
How do age and underlying conditions influence the impact of the cardiac troponin ratio in heart failure patients?
Omeone mentions ratios during dinner conversation!
We’re diving deeper now! The study found that patients with a CTR above 52.5 had a staggering 69.1% chance of encountering acute heart failure. For those under 60 years old within the same CTR range, this risk catapulted to 84.3%. On the flip side, if their CTR was lower than this threshold, the outlook brightened significantly, with a 75.4% probability of avoiding heart failure. So, CTR is indisputably the star of this show!
However, there’s more drama in different age brackets and etiologies. For patients over 60 with a high CTR, risks varied by cause. While those with infection or malignancy had a considerable chance of avoiding heart failure (69.6%), the situation turned dire for those with rheumatological issues, whose risk soared to 85.7%. An intriguing chapter indeed!
### Insights on Model Performance
Now, let’s talk about the model performances. Think of these models like clever detectives—somewhat accurate but not without their flaws. In the training set, the model managed to correctly identify about 68.1% of heart failure cases, which isn’t too shabby! But when tested on new data, it faltered a bit with a sensitivity drop to 59.7%. Take that with a grain of salt, though—every novel has its plot twists, and every model its limitations.
### Importance of Variables
When considering variable importance, CTR reigned supreme at a score of 100%! Following closely were age, NT-proBNP levels (a heart failure marker), and the underlying etiology of HPS. This hierarchy highlights the particular need to keep our eye on these key players, especially the CTR which demands continuous attention in clinical settings.
### Conclusion: A Multidimensional Approach
As we draw our narrative to a close, it’s clear that the intersection of HPS and acute heart failure is a complex puzzle—a tapestry woven with threads of ratios, age, etiology, and inflammation. This study underscores the pressing necessity for vigilant monitoring and proactive interventions. So, as we continue to explore these dimensions, understanding the role of various predictors can indeed level up our patient management strategies.
In sum, it’s about being ahead of the curve—knowing when to intervene and tailoring treatments to patient risk profiles. The journey to demystifying HPS and its heart-related intricacies continues, but with insights like these, we’re well-equipped to tackle the challenges head-on.
