The coexistence of rheumatoid arthritis and ankylosing spondylitis: Case series and literature review

Abstract

Introduction

Rheumatoid arthritis and ankylosing spondylitis are among the most common rheumatic diseases. The coexistence of both affections in a single patient is highly infrequent but seems to be underestimated in clinical practice.

Materials and methods

We described the clinical, biological, and radiological characteristics of 12 patients with concomitant rheumatoid arthritis and ankylosing spondylitis diagnosed at our hospital. The 28 disease activity score (DAS28), the Bath ankylosing spondylitis disease activity index (BASDAI), and the Ankylosing Spondylitis Disease Activity Score (ASDAS) were used as outcome measures.

Results

Twelve patients with a male-to-female ratio of 0.71 and a mean age of 62.1 ± 12.8 years were included. Rheumatoid arthritis was the first disease diagnosed in seven patients. The mean duration of rheumatoid arthritis diagnosis at the time of ankylosing spondylitis diagnosis was 20.2 ± 25 months. The first sign of ankylosing spondylitis in rheumatoid arthritis patients was incidental radiological sacroiliitis in four patients, inflammatory low back pain in three patients, and distal interphalangeal joint involvement in hands radiographs in two patients. Rheumatoid arthritis was seronegative in four patients. Erosions were observed on radiographs of the hands and/or feet in 66% of the cases and almost all the patients (11/12) had sacroiliitis on imaging studies. The mean values of the DAS28, ASDAS, and BASDAI scores at the initial diagnosis of rheumatoid arthritis/ankylosing spondylitis were 4.54 ± 1.22, 3.1 ± 0.72, and 4.1 ± 0.5, respectively.

Conclusion

The coexistence of Rheumatoid arthritis and ankylosing spondylitis is uncommon but should be considered. We could not draw a conclusion about whether the association of both disease conditions confers different characteristics.

Resumen

Introducción

La artritis reumatoide y la espondilitis anquilosante son dos de las enfermedades reumáticas más prevalentes. La coexistencia de ambas patologías en un mismo paciente es rara, pero parece estar subestimada en la práctica clínica.

Materiales y métodos

Se describen las características clínicas, biológicas y radiológicas de 12 pacientes con diagnóstico concomitante de artritis reumatoide y espondilitis anquilosante en nuestro hospital. Se emplearon como medidas de resultado la puntuación de actividad de la enfermedad 28 (DAS28), el índice de actividad de la enfermedad de la espondilitis anquilosante de Bath (BASDAI) y la puntuación de actividad de la enfermedad de la espondilitis anquilosante (ASDAS).

Resultados

Se incluyeron 12 pacientes, con una proporción hombre-mujer de 0,71 y una edad media de 62,1 ± 12,8 años. La artritis reumatoide fue el primer diagnóstico en siete pacientes. La duración media entre el diagnóstico de artritis reumatoide y el diagnóstico de espondilitis anquilosante fue de 20,2 ± 25 meses. El primer signo de espondilitis anquilosante en los pacientes con artritis reumatoide fue la sacroileítis radiológica incidental (en 4 pacientes), el dolor lumbar inflamatorio (en 3 pacientes), y la afectación de la articulación interfalángica distal en radiografías de manos en dos pacientes. La artritis reumatoide fue seronegativa en 4 pacientes. Se observaron erosiones en radiografías de manos y/o pies en el 66% de los casos, y 11 pacientes presentaban sacroileítis en los estudios de imagen. Las puntuaciones medias de DAS28, ASDAS y BASDAI al momento del diagnóstico inicial de artritis reumatoide/espondilitis anquilosante fueron 4,54 ± 1,22, 3,1 ± 0,72 y 4,1 ± 0,5, respectivamente.

Conclusiones

La coexistencia de artritis reumatoide y espondilitis anquilosante es infrecuente, pero debe ser considerada. No se pudo determinar si la asociación de ambas enfermedades confiere características clínicas distintivas.

Introduction

Rheumatoid arthritis (RA) and ankylosing spondylitis (AS) are common and severe chronic inflammatory skeletal diseases.¹ They are considered to be two independent diseases due to their clinical, genetic, pathogenic, and analytical differences.² While joint damage in RA comprises widespread destruction; damage to the spine or joint in AS is usually accompanied by ossification and ankylosis.² The pathophysiology of chronic inflammatory diseases is rooted in the interaction network of proinflammatory cytokines such as tumor necrosis factor (TNF)-α and interleukin (IL)-1, IL-6, IL-17, and IL-23.² For many patients with inflammatory conditions such as RA and spondylarthritis (SpA), their response to inhibitors of these inflammatory cytokines may differ, suggesting a disease-dependent, hierarchical cytokine effect.³ Ongoing studies implicate TNF-α, IL-1, and IL-6 in RA pathology and TNF-α, IL-17, and IL-23 in SpA.³

The typical symptoms of AS and RA are different. However, the problem arises mainly in patients with inflammatory axial complaints and seropositive symmetrical arthritis. In fact, peripheral arthritis only happens in about a third of cases in AS.⁴ They are asymmetric, involve large joints, and are rarely erosive.⁴ Furthermore, axial involvement may occur in RA patients with active and severe disease. In these cases, the overlap between RA and AS represents a main challenge for the rheumatologist, who should keep in mind the possibility of coexisting two rheumatic diseases in the same patient.

In fact, according to previous reports, the two diseases rarely coexist.^5–7 However, it seems that the phenomenon of RA and AS coexistence is more widespread than previously thought. The first case of simultaneous RA and AS was reported in 1976.⁵ Since that first report and up until now, to the best of our knowledge, 104 cases were reported in the literature.^8–10 We aimed in this retrospective study to describe the characteristics of 12 RA/AS patients and to review the previously published cases.

Methods

Patients

From January 2013 to March 2023, patients who were diagnosed at the outpatient clinic of a Tunisian tertiary rheumatology center, with RA according to the 2010 ACR/European League Against Rheumatism (EULAR) criteria,¹¹ and concomitant SA according to the Assessment of SpondyloArthritis international Society (ASAS) criteria,¹² were identified. The inclusion criterion was an age > 16 years at the time of the first chronic inflammatory rheumatic disease (CIRD). Patients with psoriasis or psoriasis family history were excluded from this study. Patients with RA associated with other connective tissue diseases were also excluded.

Clinic and laboratory data

All data regarding family history, socio-demographic characteristics, extra-articular manifestations (EAMs), past and ongoing prescribed medications (including non-steroidal anti-inflammatory drugs (NSAIDs), biologic and synthetic disease-modifying anti-rheumatic drugs (DMARDs)), were obtained from medical records. The laboratory examination included a complete blood count (CBC), erythrocyte sedimentation rate (ESR), C-reactive protein (CRP) level, rheumatoid factor (RF) positivity (with a positive cut-off value of >15 U/mL), anti-cyclic citrulline peptide (anti-CCP) (with a positive cut-off value of >10 U/mL), anti-nuclear (ANA) ELISA-test (with a positive cut-off value of >1/80 U/mL), and human leukocyte antigen-B 27 (HLA-B27) positivity.

Imaging data

A single imaging expert reviewed the imaging X-ray data of the wrists, hands, feet, sacroiliac joints (SIJs), and lungs of all the patients. Radiographs of the SIJ were scored according to the modified New York criteria (mNYC) grading system (grade 0–4).¹³ In patients with no structural according to the radiographic mNYC, MRI of the SIJs was performed with coronal, axial short TI inversion recovery (STIR), and axial T1 and T2-weighted sequences. Positive SJIs MRI was defined according to the ASAS¹¹ by the presence of bone marrow edema (BMO)/osteitis on a T2-weighted sequence sensitive for free water (such as short tau inversion recovery STIR) or bone marrow contrast enhancement on a T1-weighted sequence, inflammation must be clearly present and located in a typical anatomical area (subchondral bone) and MRI appearance must be highly suggestive of a spondyloarthritis.¹¹

Follow-up and assessment

We assessed RA activity using the disease activity score (DAS28) score¹⁴ and AS activity using the Bath ankylosing spondylitis disease activity index (BASDAI)¹⁵ and the Ankylosing Spondylitis Disease Activity Score (ASDAS).¹⁶ The three activity scores were assessed at the time of RA/AS concomitant diagnosis and after 2years of follow-up.

Statistical analysis

All statistical analyses were performed with SPSS (version 25.0; IBM, Armonk, NY). The mean ± standard deviation (SD) was used to represent the normal distribution of metrological data and the medians ± interquartile range (IQR) were used to represent non-normally distributed data. Frequencies and rates were used to represent count data.

Ethical concerns

All patients gave their consent for publishing their cases with absolute respect of anonymity.

Results

Demographic and clinical data

Twelve patients with a male-to-female ratio of 0.71 and a mean age of 62.1 ± 12.8 years were included in the study. A family history of autoimmune disease was found in three patients. Mean body mass index (BMI) was 21.4 ± 11 kg/m². Overweight (BMI ≥ 25 kg/m²) was found in four patients and obesity (BMI ≥ 30 kg/m²) in one.

RA was the first disease diagnosed in seven patients and the two CIRDs were diagnosed at the same time in two patients. The mean age at RA + AS coexistence diagnosis was 56.2 ± 16.7 years. The mean duration of RA diagnosis at the time of AS diagnosis was 20.2 ± 25 months. The first sign of AS in RA patients was incidental radiological sacroiliitis in four patients, inflammatory low back pain in three patients, and distal inter-phalangeal (DIP) joint involvement in hands radiographs in two patients. The first sign of RA in AS patients was symmetrical arthralgia (three patients).

At the time of the initial diagnosis of RA/AS, extra-articular manifestations (EAMs) were found in three (two cases of pulmonary involvement and one case of uveitis). Clinical details are summarized in Table 1.

Table 1.

Demographic and clinical characteristics of the patients with rheumatoid arthritis and concomitant ankylosing spondylitis.

RA/AS (n = 12)	Clinical data
Age ± SD (years)	62.1 ± 12.8
Gender-ratio	0.71
Family history of autoimmune disease, n	2
First CIRD diagnosed, n	RA: 7AS: 3At the same time: 2
Age ± SD at RA + AS coexistence diagnosis (years)	56.2 ± 16.7
The mean duration of RA diagnosis at the time of AS diagnosis (months)	20.2 ± 25
Extra-articular manifestations, n	3

RA = rheumatoid arthritis; AS = ankylosing spondylitis; CIRD = chronic inflammatory rheumatic disease; SD = standard deviation.

Laboratory and imaging data

Regarding laboratory findings, mean leukocyte count, hemoglobin concentration, platelet count, ESR, and CRP were 8650 ± 1340 cells/mm³, 11.3 ± 2.8 g/dL, 380.000 ± 156.000 cells/mm³, 53.3 ± 35 mm/h and 16.2 ± 14.8 mg/L respectively.

RA was seronegative in four patients. RF and anti-CCP were positive in 67% (8/12) and 58% (7/12) of cases respectively, with mean titers of 143.5 ± 108 U/mL and 114 ± 194 U/mL. ANA was positive and speckled in one patient (1/100 U/mL). HLA B27 was tested in two patients and was positive in both of them. Regarding imaging findings, 8 patients presented erosions of the hands and/or feet in their radiographs. Hip involvement was found in two patients and radiological sacroiliitis was observed in almost all patients (11/12). Among the 22 sacroiliitis observed, grade 2, grade 3, and grade 4 were observed in four, ten, and eight cases, respectively. SIJ MRI was performed in three patients and showed active sacroiliitis in all cases. Table 2 summarizes the laboratory and imaging findings of the patients.

Table 2.

Laboratory and imaging characteristics of the patients with rheumatoid arthritis and concomitant ankylosing spondylitis.

Variable	Value
CRP ± SD, Mean (mg/L)	16.2 ± 14.8
ESR ± SD, Mean (mm/h)	53.3 ± 35
RF positive, n (%)	8 (67%)
Anti-CCP positive, n (%)	7 (58%)
AAN, n (%)	1 (8%)
Positive HLA B27, n (%)	2 (16%)
Bone erosions in the wrists, hands, and feet, n (%)	8 (66%)
RA cervical spondylitis	2 (16%)
Hip involvement, n (%)	2 (16%)
Sacroiliitis in the pelvic radiograph, n (%)	11 (92%)

WBC = white blood cell count; Hb = hemoglobin; CRP = C-reactive protein; ESR = erythrocyte sedimentation rate; RF = rheumatoid factor; ACPA = anti-citrullinated protein antibodies; ANA = antinuclear antibodies.

Follow-up

For 10 patients, we are currently at a two-year decline in RA/AS diagnosis. The mean values of the DAS28, ASDAS, and BASDAI scores at the initial diagnosis of RA/AS were 4.54 ± 1.22, 3.1 ± 0.72, and 4.1 ± 0.5, respectively. After 2years of follow-up, the mean values were 3.3 ± 0.8, 2.47 ± 1.15, and 3.83 ± 0.74, respectively.

Treatment

All patients were treated with NSAIDs and methotrexate with a mean dose of 13.7 ± 3.2 mg/week. Sulfasalazine was prescribed in one patient (2 g/day). The onset of AS prompted a treatment escalation to biologic DMARDs in two RA patients with atlantoaxial subluxation. At a two-year decline in RA with concomitant AS diagnosis, tumor necrosis factor (TNF)-alpha inhibitors prescription was indicated in another patient.

Discussion

According to previous reports, RA and AS rarely coexist in the same patient^5,7 and the probability for the co-occurrence of the two diseases has been reported to range from 1/50,000 to 1/100,000.¹⁵ However, it is likely that the rarely described phenomenon of RA and AS coexistence may be more widespread than previously thought as shown in a study in 2017, where in 286,601 patients with RA, the association with other autoimmune processes was evaluated, observing that 1.16% of patients also manifested AS.¹⁷

Approximately 104 cases have been documented to date (8–10). Flores-Robles et al.⁸ carried out a literature review including the period from January 1950 to April 2020, they identified 74 cases of concomitant RA/AS and they described seven new cases from their center. Then in 2021, Zhang⁹ described 22 patients with a delayed diagnosis of either disease, and Khasru et al.¹⁰ reported a case of a 37-year-old woman with concomitant RA, AS, systemic lupus erythematosus and antiphospholipid syndrome. In the present study, we described the epidemiological, clinical, laboratory, and radiological characteristics of 12 new patients with concomitant diagnoses of RA and AS with a comparison to those of patients described in the literature (Table 3). To the best of my knowledge, this is the second literature review carried out in an exhaustive manner.

Table 3.

A literature review on the characteristics of patients with rheumatoid arthritis and concomitant ankylosing spondylitis.

	Flores-Robles et al. (2020)	Zhang et al. (2021)	Khasru et al. (2021)	Ohta et al. (2023)	Our study (2023)
Number of cases	81 (74 detected in a literature review and 7 new cases)	22	1	1	12
Concomitant diseases	RA/AS	RA/AS	RA/AS/systemic lupus erythematosus/Sjogren syndrome/antiphospholipid syndrome	RA/AS	RA/AS
Age ± SD (years)	53.7 ± 14.8	51.8 ± 18.2	37	52	62.1 ± 12.8
Male to female-ratio	2.5	0.46	1 Female	1 Female	0.71
Family history of autoimmune disease, n	9/81	NS	RA and AS in first-degree relatives	NS	2
First CIRD diagnosed, n	RA: 39/75AS: 36/75NS: 6	NS	At the same time	RA	RA: 7AS: 5At the same time: 2
Age ± SD at the first CIRD diagnosis (years)	34.9 ± 14.47	NS	NA	42	53.7 ± 16.6
Age ± SD at the second CIRD diagnosis (years)	NS	NS	NA	52	56.2 ± 16.7
The mean duration of the first CIRD at the time of diagnosis of the second	18.08 ± 13.16 years	10.5 ± 10.7 years	NA	10 years	20.2 ± 25 months
The first sign of the second disease, n	Lumbar pain: 37/76Arthritis: 35/76Uveitis: 2/76NS: 5	NS	Concomitant lumbar pain + arthritis + dry eyes + photosensivity + oral ulcer + repeated abortions	Inflammatory back pain	Incidental sacroiliitis: 4DIP joint involvement in hands radiographs: 1Inflammatory low back pain: 2Arthralgia: 3
Extra-articular manifestations (except uveitis), n	15/81	19/22 (ILD)	ILD	NS	2/12 (ILD)
Uveitis, n (%)	11/81 (13.5%)	NS	No	NS	1/12 (8.3%)
Mean WBC ± SD (cells/mm^3)	NS	8180 ± 2840	NS	6000	8650 ± 1340
Mean CRP ± SD (mg/L)	NS	67.7 ± 65.74	24	0.33	16.2 ± 14.8
Mean ESR ± SD mm/h)	NS	81.25 ± 97.2	40	NS	53.3 ± 35
RF positive, n (%)	73/79 (92.4%)NS: 2	NS	Positive RF	Positive RF	8/12 (67%)
Anti-CCP positive, n (%)	16/18 (88.88%)NS: 63	NS	Positive anti-CCP	Positive anti-CCP	7/12 (58%)
AAN, n (%)	NS	NS	Positive	NS	1/12 (8%)
Bone erosions, n (%)	65/76 (85.5%)	8/9 (88.8%)	NS	NS	8/12 (66%)
Hip involvement, n (%)	NS	Involvement of knee or hip joint: 8/22	No	NS	2/12 (16%)
Radiological sacroiliitis, n (%)	80/81 (98.7%)	2 level: 3/223 level: 17/224 level: 3/22	Grade 2 bilateral sacroiliitis	Bilateral sacroiliac joint deformation	11 (92%)
Spinal syndesmophytes, n (%)	42/75 (73.68%)	NS	NS	NS	2 (16%)

RA = rheumatoid arthritis; AS = ankylosing spondylitis; CIRD = chronic inflammatory rheumatic disease; DIP = distal interphalangeal joints; BMD = bone mineral density; ILD = interstitial Lung disease; NS = not specified; NA = not applicable; WBC = white blood cell count; Hb = hemoglobin; CRP = C-reactive protein; ESR = erythrocyte sedimentation rate; RF = rheumatoid factor; ACPA = anti-citrullinated protein antibodies; ANA = antinuclear antibodies.

In our study, the ratio of males to females was 0.71. One explanation for this could be that in 7/12 cases, the first disease was RA, an entity that mainly affects women.¹⁶ Our results go in line with those of the study of Zhang et al.⁹ where most RA/AS patients with a delay in diagnosis were middle-aged women. However, in the review of Flores-Robles,⁸ and as the first disease was AS, an entity commonly affecting men, the ratio of males to females was approximately 3 to 1.

The mean duration of the first disease up to the diagnosis of the second was 20.2 ± 25 months, relatively shorter than what whose described in the study of Zhang et al.,⁷ where the mean duration of diagnosis delay was 5.5 years. The authors explained the misdiagnosis in their study mainly by the absence of axial symptoms, patients were seldom screened for HLA-B27 antibodies or undergo an SIJ x-ray examination, and third some RA/AS patients were RF and HLA-B27-negative, which was often overlooked by clinicians.⁷ We also assumed that some administered therapy can mask the symptoms of one of the diseases, which are then subclinical.

As in the Zhang et al. study,⁷ most of our RA patients had no prominent axial complaints and the first sign of AS was an incidental radiological sacroiliitis. This suggests that the association of RA and AS might confer some different characteristics to one or both diseases and highlights that the coexistence of AS in RA patients should be suspected in the presence of radiological sacroiliitis. Nevertheless, it also should be kept in mind that sacroiliac joint inflammation may affect up to 20% of RA patients.¹⁸ The differences in the radiological image, however, are quite significant. In RA, ankylosis of sacroiliac joints is rarely observed, the changes are located in distal sections and the degree of damage is lower.¹⁹

As in previous reports,^6–8 our patients were positive for RF and anti-CCP (67% and 58%, respectively). As known, these antibodies may be positive in AS without coexisting RA. In fact, in studies by Tuossirot et al., the RF was positive in 8.3% of patients with AS and 9.8% of the control group and the difference was not statistically significant. Akar et al.²⁰ also demonstrated that a positive RF in a patient with AS does not increase the risk of peripheral joint erosion. However, anti-CCP which is occasionally present in AS, may be a marker to predict peripheral arthritis in this domain.²¹

HLA B27 was tested in two of our patients and was positive in both of them. Similarly, 89.5% of the patients in studies by Flores-Robles et al.⁷ and Zhang⁹ were positive for this antigen. HLA B27 can be positive in up to 9% of RA patients,²² but it does not increase the incidence of enthesitis or sacroiliac joint inflammation.²³ Guo et al.⁶ assumed that some environmental factors may affect the disclosure of AS in RA individuals with a positive HLA-B27.

With regards to treatment, three patients (25%) were on biologic DMARDs after two years of the RA/AS diagnosis. In the study of Flores Robles et al.,⁶ TNF-alpha inhibitors were prescribed in 10.6% of patients, while in the study of Zhang et al.,⁷ 18 of 20 patients (90%) used at least one TNF-alpha inhibitor for >6 months after 2 years of follow up. With the data presented, we could not draw a conclusion whether the association of both disease conditions increases the CsDMARDs resistance.

We assessed the DAS28, the ASDAS, and the BASDAI scores at the time of initial diagnosis of RA or AS and after 2 years. As Zhang et al. we found that after 2 years of therapy, the DAS28 and BASDAI scores of RA/AS patients were lower than the first assessment. However, this finding should therefore be taken with precaution, since all patients were on DMARDs therapy.⁷

We admitted some limits of our study, mainly the small number, the retrospective method design, and the fact that much of the published data is not very precise which makes them difficult to analyze and, therefore, draw more precise conclusions.

Conclusion

In conclusion, the coexistence of RA and AS in the same patient is not rare as in the clinical setting, but it is often unrecognized. Particular attention should be paid to RA patients with symptoms of typical axial SpA, and to AS patients with peripheral joint involvement. It should be kept in mind that one diagnosis does not necessarily exclude another. Longitudinal or prospective studies are needed for the diagnosis and treatment of RA/AS patients.

Ethical considerations

None.

Funding

None.

Conflict of interest

None.

Acknowledgment

None.