The main objective of this study is to develop a diagnostic model with high sensitivity and specificity to help pediatricians to recognize malignancy and minimalize unnecessary referral in children with lymphadenopathy. Moreover, we will perform a…
ID
Source
Brief title
Condition
- Lymphomas Hodgkin's disease
- Lymphomas non-Hodgkin's unspecified histology
- Spleen, lymphatic and reticuloendothelial system disorders
Synonym
Research involving
Sponsors and support
Intervention
- No intervention
N.a.
Outcome measures
Primary outcome
<p>The primary endpoint of this research is a machine learning diagnostic model for malignant lymphoma with a high specificity (85%+) for a minimum sensitivity (95%). The malignant lymphoma diagnosis will be confirmed by biopsy or excluded if there is an alternative diagnosis or spontaneous regression of the lymph node.</p>
Secondary outcome
<p>Additionally, we will make a separate model for a specific type of lymphoma (Nodular lymphocyte-predominant Hodgkin lymphoma). The third endpoint is a cost-effectiveness analysis. We will compare the costs and effects of the diagnostics tests using our model with standard care.</p>
Background summary
Lymphadenopathy is common in children and is mostly caused by benign conditions. However, it can indicate a serious underlying disease, such as malignant lymphoma. Physical examination, standard laboratory and radiology tests are not always sufficient to make the correct diagnosis. Sometimes a lymph node biopsy is needed to obtain a definitive diagnosis. Despite being rare, there is always a risk of complications from these procedures, such as bleeding or scarring. The challenge for healthcare providers is to avoid unnecessary aggressive evaluation and biopsies while ensuring timely and accurate diagnoses for children with severe underlying diseases.
Study objective
The main objective of this study is to develop a diagnostic model with high sensitivity and specificity to help pediatricians to recognize malignancy and minimalize unnecessary referral in children with lymphadenopathy. Moreover, we will perform a cost-effectiveness analysis to determine whether this model will result in cost reduction in the diagnostic process of children with lymphadenopathy.
Study design
We will perform a multi-center, prospective cohort study in different hospitals in the Netherlands. We will collect data in a standardized way from outpatient clinic, emergency room (ER) visits and and hospital admissions. Based on a flowchart pediatricians will choose which additional test are necessary. We will conduct univariate analyses for various variables related to medical history, physical examination and additional tests, to provide more clarity on predictive factors for lymphoma in children with lymphadenopathy. In addition, we will perform multivariate analysis to create a machine-learning logistic regression model.
On top of that, we will ask the patient and their parent(s) or caregiver(s) to complete a short questionnaire about their quality of life and how this hospital visit has affected their daily life. We will use this information for the cost-effectiveness analysis.
Intervention
Not applicable.
Study burden and risks
If additional blood tests are needed, an extra analysis will be done on the back-up serum sample. This means that patients do not need to have their blood drawn twice and do not undergo any additional risk.
The questionnaires patients and parents are asked to fill out will be short and the questions are not considered intrusive.
A. Beishuizen
Heidelberglaan 25
Utrecht 3584 CS
Netherlands
088-972 72 72
A.Beishuizen-2@prinsesmaximacentrum.nl
A. Beishuizen
Heidelberglaan 25
Utrecht 3584 CS
Netherlands
088-972 72 72
A.Beishuizen-2@prinsesmaximacentrum.nl
Trial sites in the Netherlands
Listed location countries
Age
Inclusion criteria
- Reason of referral /hospital admission: lymphadenopathy
- Informed consent by patient and/or parent(s)/guardian(s)
Exclusion criteria
- Gland(s) < 1 cm (exception: supraclavicular lymph nodes)
- Gland(s) spontaneously in regression at the time of first visit
- Clinical picture is consistent with lymphadenitis
- Swelling appears not to be a lymph node (for example, branchial cleft cyst or haemangioma)
Design
Recruitment
Medical products/devices used
IPD sharing statement
Plan description
Followed up by the following (possibly more current) registration
No registrations found.
Other (possibly less up-to-date) registrations in this register
No registrations found.
In other registers
| Register | ID |
|---|---|
| Research portal | NL-009186 |