Eurofever Project - The Diagnostic score for Periodic fever


How to calculate the Diagnostic Score

Step 1. Download the Clinical Chart (Appendix 1)
Step 2. Fill in the Clinical Chart, possibly  while you are visiting the child. Ask directly to the parents the presence and the frequency of each clinical manifestation associated with the fever attacks. It will  takes no more than 5-10 minutes.
Step 3. Insert the correct answer for each requested item
Step 4. Now you can calculate the score. Three different results will appear: 1) the score, 2) the probability to be positive, 3)  the predicted group (low or high risk)

Age at onset (months)
Abdominal pain
Aphtosis
Thoracic pain
Diarrhea
Family history
  
Calculate

A diagnostic Flow-chart for children with suspected monogenic Periodic Fever
The  Diagnostic score and  the diagnostic flow chart  (see the Figure) are aimed to provide an evidence-based tool to  General Pediatricians that usually observe for the first time children with periodic fevers. Its  aim is to timely select those patients at higher risk to carry mutations of known genes. These patients should be promptly referred to Specialized centers that will provide the molecular analysis and the proper clinical expertise.  
Based on our study  we propose the following diagnostic  flow-chart:


Click on the image to enlarge

  1. Include patients highly suspected for monogenic Periodic Fever (see Inclusion and Exclusion Criteria)

  2. Calculate the Diagnostic Score,  as above.
    - In patients at high risk to be positive at the genetic test, the molecular analysis can be driven according to the indications of the regression tree analysis (see the Flow-chart).
    In the present study most of the patients were of west European Caucasian  origin thus,  the multivariate analysis was not able to include the ethnic origin of the patients among the possible discriminating variables.  However, it is conceivable that MEFV gene  should be considered as the first gene to be screened in patients belonging to populations with an high prevalence of Familial Mediterrean Fever,  as for Jews, Arabs, Armenians and Turks   fulfilling the current clinical FMF Criteria  (see Flow chart).
    - Patients at low risk should not undergo directly  molecular analysis. These patients should be followed longitudinally in order to disclose possible new clinical manifestations or to identify those children undergoing a spontaneous reduction of the frequency and severity of fever episodes. The possibility to longitudinally re-evaluate a patient may allow the detection of those genetically positive patients excluded from the first screening phase and that could be sent to Specialized centers in a second time. 

IMPORTANT:
Please note that the Diagnostic score was established and validated on a pediatric population, with a mean age at disease onset of 4.3 years. Thus, the accuracy of the Score in patients with a disease onset after the first decade has not been evaluated.
In our study the cut-off of 1.32 for the score (or 15% probability to be positive) display  the best accuracy either in the Training  (sensitivity = 95%, specificity = 82%) and in  the Validation (sensitivity = 87%, specificity = 72%)  groups of patients.
In order to achieve an higher sensitivity, you can choose to consider  a lower cut-off (i.e. 10% of probability to be positive). Of course you will increase the probability to have an higher number of negative genetic tests.