The ACHILES study was planned as 3 independent cohorts of patients (SpA, psoriasis and IBD), defined by the main diagnosis at the time of enrollment. The study was conducted in 15 Spanish hospitals with the participation of three departments: Rheumatology, Dermatology and Gastroenterology. Each department included patients aged ≥18 years diagnosed with one of the following IMID: SpA (including ankylosing spondylitis, psoriatic arthritis, undifferentiated spondyloarthritis, enteropathic arthritis or other), psoriasis and IBD, thus constituting the 3 independent cohorts. Patients with 2 or more diseases followed by 2 or more specialists could only be included into one of the cohorts. Patients were followed up for a period of 2 years by the same specialist who included them in the study.

The objective of this work is to describe the baseline characteristics of the cohort of patients with a diagnosis of SpA at the time of inclusion into the study and the prevalence of psoriasis, uveitis and IBD (primary endpoint). These patients were recruited from the rheumatology clinics of the participating hospitals.

In this cohort study researchers included adult patients (age ≥18 years) with an established or de novo diagnosis of SpA. The diagnosis of SpA was made in all cases by a rheumatologist. Patients who, in the opinion of the investigator, presented any circumstance that would prevent their proper follow up for 2 years were excluded. The protocol was approved by the Ethics Committees of the participating centers and was conducted in compliance with the standards of good clinical practice.

Researchers could include patients with a previous diagnosis of SpA by random assignment or consecutively. In the first case, a list of patients with SpA who had been seen in rheumatology in the 6 past months prior to study start was generated, and on that basis patients were randomly selected. In the SpA cohort, however, most patients already diagnosed (∼90%) were included consecutively. Patients with a new diagnosis of SpA (seen de novo in Rheumatology) who met criteria were included consecutively into the study.

The inclusion period lasted from March 2008 until December 2010. The information was gathered from personal interviews with the patient and review of the clinical history, following a protocol agreed upon by the 3 specialties. The diagnosis of each disease was made on the basis of the diagnoses listed in the medical record until the time of study entry and to determine prevalence at baseline, the diseases associated with flares (Uveitis) were counted as a case not only when they were present at baseline, but also when the patient presented a flare in the past, reflected as such in the medical record. In the study no laboratory test or radiological test was performed and baseline data and diagnoses were collected from medical records.

The sample size calculation was based on the primary endpoint, i.e. the prevalence of these diseases in each of the cohorts. For each cohort of patients (SpA, IBD and psoriasis), an expected prevalence for each disease of 3%, a confidence level of 95% and an accuracy of 1% was assumed. The sample size was calculated, for each cohort, at 1075 patients. Recruitment was lower than expected and the final sample of the SpA cohort (601 patients) allows an accuracy of 3% for the prevalence found.

Continuous variables were described as mean±standard deviations, and medians and interquartile ranges (IQR). Categorical variables were presented using frequencies and percentages. The population was divided according to various demographic and clinical variables to assess differences in the prevalence of various diseases. These differences were analyzed with the chi-square test. The Student's t test of for independent groups was used to compare means of the 2 groups. When categorical variables were related, the proof of the chi-square test was used. Subsequently, a logistic regression model was adjusted to study what demographic and clinical variables were independently associated with the coexistence of uveitis, psoriasis and IBD, presenting odds ratios (OR) and 95% confidence interval (CI). In all the contrasts, the null hypothesis was rejected when the alpha error was less than 0.05. For processing data, the SPSS 15.0 software package was employed.

601 patients with SpA were included in the study, 379 men (63.1%) and 222 women (36.9%). Their mean age±SD was 47.9±12.9 years. Only 79 patients (13.1%) were newly diagnosed; the majority (86.9%) had an established diagnosis of SpA. In the latter, the median duration of disease was 8.7 years (IQR 25–75: 3.3–16.5). Baseline diagnoses of SpA were: ankylosing spondylitis: 55.1% (n=332), psoriatic arthritis: 25.1% (n=151), undifferentiated spondyloarthritis: 16.1% (n=97), enteropathic arthritis: 2.5% (n=15) and other 1.3% (n=8). 15% presented extra-articular manifestations associated to SpA (other than those studied [uveitis, psoriasis or IBD], mainly conjunctivitis [5.0%], cystitis [2.3%], nail hyperkeratosis [6.0%], balanitis [1.5%] and a lower prevalence of 1%, pulmonary fibrosis, aortic and renal amyloidosis). 19.5% had a family history of SpA. The most frequent comorbidities were depression (8.3%), anemia (5.3%), dermatitis and other skin lesions (8.0%) as well as cardiovascular risk factors (obesity [20.4%], smoking [26.6%], hypertension [22.1%], hypercholesterolemia [16.0%] and diabetes mellitus [6.0%]).

In the SpA cohort, 262 patients had at least one of these IMID, with a prevalence of 43.6% (95% CI, 39.7–47.6). The most common IMID was psoriasis, present in 167 patients (prevalence: 27.8%, 95% CI, 24.4–31.5); most of them showed plaque psoriasis (88.6%). A total of 82 patients had uveitis (prevalence: 13.6%, 95% CI, 11.1–16.6). Finally, 31 patients were diagnosed with IBD (prevalence: 5.1% [95% CI, 3.7–7.2]), with the following distribution: Crohn's disease: 18 patients (3.0%, 95% CI, 1.9–4.7), ulcerative colitis, 6 patients (1.0%, 95% CI, 0.5–2.2) and undifferentiated colitis: 7 patients (1.2%, 95% CI, 0.6–2.4). In 18 patients (3.0%; 95% CI, 1.9–4.7), 2 IMID coexisted concomitantly to SpA.

The prevalence of the major diagnostics is presented in Table 1. In patients with ankylosing spondylitis, uveitis was the most common extraarticular IMID (prevalence 19.0% [95% CI, 15.1–23.5]). Virtually all patients with psoriatic arthritis had psoriasis (prevalence 94.7% [95% CI, 89.9–97.3]). When we excluded patients with psoriatic arthritis, the prevalence of psoriasis in patients with other types of SpA was 5.6% (95% CI, 3.8–8.1). Moreover, 100% of the patients with enteropathic arthritis presented IBD and, by excluding patients with enteropathic arthritis, the prevalence of IBD in patients with the rest of SpA was 2.7% (95% CI, 1.7–4.4) (Table 1).

The chronology of diagnoses is described in Table 2. The diagnosis of IBD was prior to that of SpA in 50% of the cases, with a median of 10 years earlier. In 60% of the patients with psoriasis, this was diagnosed before SpA (median 10.5 years earlier) and only 14% had been diagnosed later. 39% of the uveitis were diagnosed prior to SpA (median: 4.5 years earlier).

Prevalence in different subgroups. The prevalence of each diagnosis per subgroup is shown in Table 3. Overall, the prevalence was numerically higher in women due to a significantly higher prevalence of psoriasis (32.9% vs 24.8% in women in men, P=.033). No differences were seen in patients with a diagnosis of established or de novo SpA. The overall prevalence of uveitis, psoriasis and IBD was also significantly higher in patients with the presence of other extraarticular manifestations associated to SpA different than those studied (65.6% versus 39.7% in patients without other extra-articular manifestations associated to SpA, P<.001). The prevalence of IBD was similar across subgroups analyzed. Psoriasis was more prevalent in women, in patients with no family history of SpA and patients with other extra-articular manifestations associated with SpA. Finally, uveitis was more frequent in patients with a longer history of the disease and in patients with a family history of SpA (Table 3).

Multivariate analysis. To assess the variables related to the presence of the 3 diseases combined, multivariate analysis that included the variables age, gender, prior or de novo diagnosis, family history of SpA, duration (<4 years, 4–8 years, >8 years was performed) and the presence of extra-articular manifestations associated with SpA. In the resulting model, the overall prevalence of IMID is independently associated with age (OR=increase of 1.02 per year of age), female gender (OR=1.4 with respect to males) and the presence of other extra-articular manifestations associated to SpA different to those studied (OR=2.8) (Table 4). Models were later performed for each of the diseases studied. IBD is associated with the absence of family history of SpA. Psoriasis is associated with age, female gender, absence of family history of SpA, presence of extra-articular manifestations associated with SpA and shorter evolution of SpA. Finally, uveitis was associated independently with a greater time since onset of SpA and family history of SpA.

The authors state that the study procedures conformed to the ethical standards of the committee responsible for human experimentation and were in accordance with the World Medical Association and the Declaration of Helsinki.

The authors declare that they have followed the protocols of their workplace regarding the publication of data from patients, and all patients included in the study have received sufficient information and gave written informed consent to participate in the study.

The authors state that no patient data appears in this article.