Trajectories of Biological Values and Vital Parameters: An Observational Cohort Study of Adult Patients with Sickle Cell Disease Hospitalized for a Non-Complicated Vaso-Occlusive Crisis

Raphaël Veil, Simon Bussy, Vincent Looten, Jean‐Benoît Arlet, J. Pouchot, Anne‐Sophie Jannot, Brigitte RanqueJournal of Clinical Medicine 2019 · DOI

TL;DR

This study establishes typical trajectories for key biological markers and vital signs in adult patients experiencing a non-complicated vaso-occlusive crisis (VOC) of sickle cell disease (SCD). By analyzing longitudinal data from a clinical data warehouse, the researchers provide a baseline for what to expect during a VOC hospitalization, which can help clinicians identify deviations that might signal complications or misdiagnoses.

What Problem Does This Paper Address?

Sickle cell disease (SCD) is a common genetic blood disorder that can lead to painful vaso-occlusive crises (VOCs). While VOCs are the primary reason for hospitalization in SCD patients, the expected changes in routine biological markers and vital signs during a non-complicated VOC have not been well-defined. This lack of established "normal" trajectories makes it challenging for clinicians to:

  • Diagnose VOC accurately: Differentiating VOC pain from other pain sources can be difficult.
  • Monitor patient progress: It's hard to know if a patient is improving as expected or if a complication is developing.
  • Identify complications early: Deviations from expected trends could be the first sign of infection, acute chest syndrome, or other issues.

This paper aims to fill this knowledge gap by characterizing the typical temporal patterns of common laboratory results and vital signs during a standard VOC hospitalization.

Methods

The researchers employed a novel spline-based approach to analyze longitudinal data extracted from a clinical data warehouse (CDW). This method is designed to handle time-dependent variables that are measured at irregular intervals for different patients.

Here's a breakdown of the key methodological aspects:

  • Data Source: A clinical data warehouse (CDW) from Georges Pompidou University Hospital (GPUH) in Paris, France, was used. This CDW aggregates routine care data from electronic health records (EHRs).
  • Data Extraction: Data included demographics, vital signs, diagnoses (ICD-10 codes), laboratory results, and free-text reports. Specialized tools were used to extract information from unstructured text.
  • Spline-Based Trajectory Analysis: The core of their method involved:
  • Creating a uniform time grid for each patient's hospital stay.
  • Fitting smoothing splines to individual patient trajectories for each variable.
  • Computing an average trajectory with a 95% confidence interval from these individual splines. This approach allows for visualization of trends even when data points are sparse or unevenly distributed.
  • Stratifying analyses by sex where relevant.
  • Threshold Analysis: For selected variables, they also calculated the proportion of patients whose values fell above or below specific clinically relevant thresholds at different time points during the stay.

Datasets

The study utilized a retrospective cohort of 329 hospital stays for non-complicated VOCs from adult patients admitted to the internal medicine department of Georges Pompidou University Hospital (GPUH) between January 1, 2010, and December 31, 2015. This represents real-world clinical data collected as part of routine patient care.

Key Insights and Ideas

  • VOCs have predictable, yet dynamic, biomarker profiles: While VOCs are often characterized by pain, the underlying biological changes follow discernible patterns over time, not just at admission.
  • Hemoglobin is a slow mover: Hemoglobin levels tend to decrease during a VOC and rarely recover to baseline levels before discharge, highlighting the prolonged impact of the crisis on red blood cell dynamics.
  • White blood cells react quickly, then subside: Leukocyte counts spike upon admission, likely due to the inflammatory response, but then rapidly decrease, suggesting the initial inflammatory surge is managed.
  • Eosinophils are a surprise player: Unlike other white blood cells, eosinophil counts show a slow but steady increase throughout the first five days of hospitalization, a trend that warrants further investigation.
  • CRP is a key indicator of inflammation and resolution: C-reactive protein (CRP) rises early and stays elevated, only beginning a slow decline towards the end of the stay. This provides a marker for ongoing inflammation and a potential indicator of resolution.
  • Temperature is generally stable: For non-complicated VOCs, fever above 38°C is uncommon, suggesting that persistent high temperatures might point towards complications.
  • A new tool for longitudinal data: The development and application of the spline-based trajectory analysis method is a significant contribution, offering a robust way to visualize and understand time-series data from clinical settings.

Key Results

  • Hemoglobin: Decreased from an average of 9.5 g/dL at admission to 8.5 g/dL by day 5. Steady-state hemoglobin was around 9 g/dL.
  • White Blood Cells (WBC): Showed an initial spike above 12.10⁹/L at admission, stabilizing around 10.10⁹/L after day 2.
  • Neutrophils: Followed a similar trend to overall WBC count.
  • Eosinophils: Increased from 2.10⁸/L at admission to 4.10⁸/L around day 5.
  • C-Reactive Protein (CRP): Rapidly increased in the first 2 days, stabilizing around 60 mg/L before slowly decreasing. Crucially, over 95% of non-complicated VOC stays had CRP below 100 mg/L within the first 24 hours, and above normal levels after 48 hours.
  • Temperature: Remained stable around 37°C throughout the stay. Less than 5% of patients exceeded 38°C during their entire stay.
  • Platelets: Remained stable between 300-350.10⁹/L.
  • LDH and Bilirubin: Showed mild decreases during the stay.

Limitations and Future Work

The authors acknowledge several limitations:

  • No Control Group: The study focused on non-complicated VOCs, making it difficult to perform sensitivity/specificity or comparative analyses against a true control group (e.g., patients with unrelated pain).
  • Diagnostic Bias: The exclusion criteria for non-VOC pain and complicated VOCs might have inadvertently introduced bias. For instance, isolated fever, which can be a sign of complication, was not used as an exclusion criterion, but the definition of complicated VOCs was strict.
  • Single-Center Study: The findings are from a single SCD referral center, which may have specific patient populations and care practices. This limits generalizability and calls for reproducibility studies.
  • Expert Bias: The monitoring practices might have been influenced by the expertise of the center, potentially leading to measurement bias compared to less specialized centers.
  • Pre-Admission Data: Lack of detailed data on self-treatment prior to seeking medical care limits the ability to assess its impact on measured variables.

Future work suggested includes:

  • Multicenter Studies: Conducting similar studies across multiple centers with varying levels of expertise would strengthen the findings and confirm reproducibility.
  • Broader Data Integration: Leveraging EHR and CDW data from more diverse settings could provide a more comprehensive understanding.
  • Further Investigation of Eosinophils: The distinct trajectory of eosinophils warrants deeper exploration into their role in non-complicated VOCs and potential links to adrenal function or other mechanisms.

Summary updated

Abstract

Hospital admission of patients with sickle-cell disease (SCD) presenting with a vaso-occlusive crisis (VOC) can be justified by pain refractory to usual outpatient care and/or the occurrence of a complication. Yet, the trajectories of vital parameters and standard biomarkers throughout a non-complicated VOC has not been established. In this observational cohort study, we describe the course of routine parameters throughout 329 hospital stays for non-complicated VOC. We used a new spline-based approach to study and visualize non-specific time-dependent variables extracted from the hospital clinical data warehouse. We identified distinct trends during the VOC for hemoglobin level, leukocytes count, C-Reactive Protein (CRP) level and temperature. Hemoglobin decreased after admission and rarely returned to steady state levels before discharge. White blood cell counts were elevated at admission before immediately decreasing, whereas eosinophils increased slowly throughout the first five days of the stay. In over 95% of non-complicated VOC-related stays, the CRP value was below 100 mg/L within the first day following admission and above normal after 48 hours, and the temperature was below 38 °C throughout the entire stay. Knowing the typical trajectories of these routine parameters during non-complicated VOC may urge the clinicians to be more vigilant in case of deviation from these patterns.
BibTeX
@article{veil2019trajectories,
  author = {Raphaël Veil and Simon Bussy and Vincent Looten and Jean‐Benoît Arlet and J. Pouchot and Anne‐Sophie Jannot and Brigitte Ranque},
  title = {{Trajectories of Biological Values and Vital Parameters: An Observational Cohort Study of Adult Patients with Sickle Cell Disease Hospitalized for a Non-Complicated Vaso-Occlusive Crisis}},
  journal = {Journal of Clinical Medicine},
  year = {2019},
  doi = {10.3390/jcm8091502},
}
Figure