Article Text

FRI0557 Estimation for the remision cutoff point for hospital universitario de la princesa index (HUPI) in a longitudinal cohort of arthritis of recent onset. validation in a long-term arthritis register.
  1. L. Sala1,
  2. I. Castrejón2,
  3. L. Carmona3,
  4. A. Ortiz1,
  5. E. Toledano4,
  6. S. Castañeda1,
  7. A. García Vadillo1,
  8. I. González Álvaro1
  1. 1Rheumatology department, Hospital Universitario de La Princesa, Madrid, Spain
  2. 2Division of Rheumatology, NYU Hospital for Joint Diseases, New York, United States
  3. 3Escuela de Ciencias de la Salud, Universidad Camilo José Cela
  4. 4Rheumatology department, Hospital Clínico San Carlos, Madrid, Spain


Background Our group has recently described a new index to assess disease activity in patients with early arthritis (EA). The Hospital de Universitario de la Princesa Index (HUPI) is easy to calculate and avoids some biases existing in currently validated indexes. The HUPI is the simple sum of the scores assigned to each of the variables included in the DAS (count of swollen and painful joints, global assessment of the patient’s disease and acute phase reactants) according to their interquartile distribution in our EA population (0 to 3 according to the quartile). The index allows the use of ESR and / or CRP as available. We have previously shown that HUPI is as accurate as DAS28 or SDAI, with the advantage to have greater sensitivity to change1

Objectives To estimate the cutoff point for remission in HUPI and validate it in a long term RA cohort

Methods We used clinical data from 568 visits in 207 patients from our EA register. Three definitions of remission were used: Boolean and SDAI 2011 ACR/EULAR criteria, as well as that established by six experienced rheumatologists by consensus. In addition, the six rheumatologist established the low, moderate and high disease activity levels in the EA cohort. We also used EMECAR (a national cohort of rheumatoid arthritis consisting in 2000 visits in 789 patients randomly selected from 34 centers in Spain) in which remission was define by Pinals et al preliminary criteria. The cutoff points were selected considering the best combination of sensitivity and specificity for each of them using ROC analysis.

Results The best cutoff point for remission in the EA population was 2.5 with the two ACR/EULAR criteria SDAI <3’3 (Specificity 90%, Sensitivity 94%) or Boolean (Sp 84%, Se 95%), as well as for the expert opinion (S e89%, Sp 89%). It was also valid for the long term RA EMECAR cohort (using Pinals criteria: E 49%, S 95%). The remaining cutoff points were 5 to discriminate between mild and moderate activity (S 67%, E 92%) and 9 to discriminate between moderate and high (S 93%, E 89%). The area under the curve was higher in value cutoffs for remission and high activity than for the point which discriminates between low and intermediate.

Conclusions The cutoff point for remission in HUPI show high sensitivity and specificity in different populations.


  1. - Castrejon I et al. Development and validation of a new disease activity index as a numerical sum of 4 variables in patients with early arthritis. Arthritis Care Res (Hoboken). 2012. PMID: 23002022

Acknowledgements This study was supported with funding from FIS ISCIII PI11/00551, PI11/00505 and RETICS RD08/0075 (RIER).

Disclosure of Interest None Declared

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.