Article Text
PDF
Abstract
Objectives Chronic kidney disease (CKD) is a comorbidity in psoriatic arthritis (PsA). We aimed to define the prevalence of CKD in patients with PsA, describe their long-term renal outcomes and identify risk factors for CKD development.
Methods We included patients with PsA followed by our prospective observational cohort. We defined CKD as an estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m2 for at least 3 months. We characterised long-term renal outcomes of CKD cases identified following clinic entry. We used time-dependent Cox regression models to identify factors associated with CKD development.
Results Of 1336 patients included in the study, 123 (9.2%) had CKD. Of these, 25 (20.3%) were observed to have CKD at clinic entry and 98 (79.7%) developed CKD during follow-up at a median (IQR) of 8.2 (2.8–14.0) years from baseline. Doubling of baseline creatinine was observed in 18 of 98 (18.3%) new patients with CKD. 49 (50%) patients developed a sustained ≥40% reduction in baseline eGFR. Two patients developed eGFR <15 mL/min/1.73 m2. In the multivariate Cox regression model adjusted for age at study entry, sex and baseline eGFR, factors independently associated with the development of CKD included diabetes mellitus (HR 2.58, p<0.001), kidney stones (HR 2.14, p=0.01), radiographic damaged joint count (HR 1.02, p=0.02), uric acid (HR 1.21, p<0.001; 50-unit increase), daily use of non-steroidal anti-inflammatory drugs (NSAIDs) (HR 1.77, p=0.02) and methotrexate use (HR 0.51, p=0.01).
Conclusion CKD is not infrequent in PsA. Its development is associated with related comorbidities, joint damage and NSAID use. Methotrexate seems to be protective.
- Psoriatic Arthritis
- Methotrexate
- Anti-Inflammatory Agents, Non-Steroidal
Data availability statement
Data are available on reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Patients with psoriatic arthritis (PsA) commonly have comorbidities that predispose to visceral organ damage, including kidney disease. The latter remains inadequately characterised.
WHAT THIS STUDY ADDS
Chronic kidney disease (CKD) is not infrequent in patients with PsA.
Comorbidities, joint damage and daily use of non-steroidal anti-inflammatory drugs are all independently associated with CKD development in PsA.
Methotrexate, commonly introduced early in the disease course, may provide a renoprotective effect.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Control of risk factors and appropriate use of medications in PsA can help preserve renal function.
Introduction
Psoriatic arthritis (PsA) develops in up to 30% of patients with psoriasis and is characterised by diverse clinical features.1 It is recognised that PsA often leads to impaired function and a reduced quality of life.1 The disease is associated with multiple comorbidities, including obesity, type 2 diabetes mellitus (DM), hypertension, metabolic syndrome, fatty liver and an increased risk of cardiovascular events. Despite the paucity of data on PsA, kidney disease has also been described.2
Several studies have shown an increased incidence of chronic kidney disease (CKD), and even glomerulonephritis and end-stage kidney disease (ESKD), in patients with psoriasis,3–9 especially in those with concomitant arthritis10–12 and severe skin disease.5 8 9 This not only includes a decline in kidney function but also albuminuria.13 14 In an epidemiological mendelian randomisation study, Tan et al15 concluded that psoriasis is causally associated with a lower level of estimated glomerular filtration rate (eGFR), whereas renal function has no reverse causality on psoriasis. Systemic inflammation, immune dysfunction, comorbidities (eg, hypertension, DM and atherosclerosis) and nephrotoxic medications (especially non-steroidal anti-inflammatory drugs (NSAIDs)) are all contributing factors to renal dysfunction in psoriatic disease.12
Renal abnormalities in PsA include low eGFR, abnormal urinary findings (such as microalbuminuria), IgA nephropathy, membranous glomerulonephritis and secondary amyloidosis.16 The prevalence of CKD in PsA patients (extracted from a psoriasis cohort) was 18.8% in one study.10 In another analysis, 16% of patients with ‘seronegative arthritis’ (including PsA) were observed to have an eGFR of less than 60 mL/min/1.73 m2.16 23% of PsA patients had laboratory renal abnormalities in a different cohort.17 Despite these data, no previous study has adequately characterised long-term renal outcomes and CKD risk factors in patients with PsA.
This study aimed to define the prevalence of CKD in patients with PsA, describe their long-term renal outcomes and identify the risk factors associated with CKD development.
Patients and methods
Setting
The University of Toronto PsA clinic has recruited and prospectively followed patients with PsA since 1978. Patients are enrolled if they have psoriasis and inflammatory arthritis. Over 99% fulfil the ClASsification criteria for Psoriatic ARthritis.18 Patients are evaluated at the time of recruitment into the clinic (baseline) and every 6–12 months according to a standard protocol that includes a detailed history, physical examination and laboratory assessment. Radiographs of peripheral joints and spine are obtained at baseline and every 2 years and are scored according to the modified Steinbrocker method for peripheral joints19 and the New York criteria for sacroiliac joints.20 The presence of syndesmophytes is also recorded. Radiographs are scored by at least two rheumatologists by consensus. All information collected is stored on a web-based database. Of note, we have previously shown that the methods of clinical and radiographic evaluation in our cohort are reliable, with no systematic bias with regard to disease severity and follow-up.21
Patient selection
We retrieved data from patients enrolled in the cohort between January 1978 and July 2023.
Definition of CKD
We defined CKD as an eGFR of less than 60 mL/min/1.73 m222 on two consecutive visits at least 3 months apart during follow-up (counting the first visit as the event date). We based the calculation of eGFR on the CKD Epidemiology Collaboration equation.23
Long-term renal outcomes of the new CKD cases
We defined three adverse renal outcomes during follow-up: doubling of baseline creatinine, a sustained ≥40% reduction in baseline eGFR and eGFR <15 mL/min/1.73m2. These outcomes are accepted surrogates of progression to or current ESKD.24–26 We mandated the presence of each outcome on at least two consecutive visits (at least 3 months apart) to be counted. We also assessed the time to the outcome.
Data collection
From our programme database, we retrieved data at the patients’ first evaluation in the clinic and at 6–12 months intervals, including age, sex, ethnicity, year of clinic entry, age at the diagnosis of psoriasis and PsA, disease duration, smoking status, body mass index (BMI), and comorbidities (hypertension, DM, kidney stones, and ischaemic heart disease (IHD)). We summarised musculoskeletal disease activity using the Disease Activity Index for Psoriatic Arthritis score.27 We assessed skin disease activity using the Psoriasis Area and Severity Index. We also retrieved other important disease-related measures including the presence of nail disease, enthesitis and dactylitis. Laboratory data included creatinine, eGFR, cholesterol, triglycerides, uric acid, erythrocyte sedimentation rate (ESR), C reactive protein (CRP), antinuclear antibody and urinalysis. Elevation of CRP was defined as a level >3 mg/L. Abnormal ESR was considered in males when ≥15 mm/hour if age <50 years and ≥20 mm/hour if age ≥50 years, and in females when ≥20 mm/hour if age <50 years and ≥30 mm/hour if age ≥50 years. We included the following radiographic features (obtained at 2-year intervals): syndesmophytes, sacroiliitis and the radiographic damaged joint count (RDJC) (the number of joints with erosions). We also retrieved information on past and current treatments with NSAIDs, conventional synthetic disease-modifying antirheumatic drugs (DMARDs) (methotrexate (MTX), sulfasalazine and leflunomide), targeted synthetic DMARDs (apremilast, tofacitinib and upadacitinib) and biological DMARDs (tumour necrosis factor inhibitors (TNFi), interleukin (IL)-17i, and IL-23i (including IL-12/IL-23i)).
Statistical analysis
We present descriptive statistics for continuous variables according to the mean (SD) if normally distributed or median (with IQR) otherwise, and count data according to frequencies and percentages. We used the t-test for hypothesis testing regarding normal continuous variables and Pearson’s χ2 tests for categorical variables. Univariate Cox regression models were used to identify factors associated with time to the development of CKD. Models were fitted for all variables of interest, adjusting for age, sex and eGFR at baseline. We used the time of clinic entry (baseline) as the time origin; patients who developed CKD before clinic entry were removed from the predictive analysis. For time-varying covariates, values from the most recently completed assessment were carried forward until a new covariate value became available. All covariates, except age, sex and eGFR at baseline, were considered as time-varying covariates. Variables with a p<0.1 in a univariate model, along with others chosen based on clinical significance, were included in a multivariate Cox regression model28 also adjusting for age, sex and eGFR at baseline. A reduced model was then obtained through backward elimination of the full model. All analyses were conducted by using R V.4.2.3.
Results
Prevalence of CKD in patients with PSA
Of the 1336 patients included in this study, 25 (1.9%) were observed to have CKD at their first clinic visit; 98 (7.3%) developed it during follow-up at a median of 8.2 (2.8, 14.0) years from clinic entry. In total, 123 (9.2%) patients had CKD.
Patient characteristics
The mean (SD) age of our cohort was 44.9 (13.1) years; 746 (55.8%) patients were males and 1149 (86.0%) were white. Most patients (840 (62.9%)) were recruited since the year 2000. The mean creatinine level was 78.7 (35.4) μmol/L and the mean eGFR level was 98 (19.9) mL/min/1.73 m2. Characteristics of the CKD cases as compared with non-CKD cases at the first evaluation in the clinic are shown in table 1.
Baseline demographic, clinical and radiographic characteristics, as well as treatment regimens of patients with CKD and those without CKD at the time of clinic entry
Patients with CKD were older at the baseline visit and at the time of diagnosis of psoriatic disease. At baseline, they had a longer disease duration. They also had more comorbidities, including DM, hypertension, kidney stones and IHD. There were no differences in the swollen joint count and tender joint count, but RDJC was higher in the CKD group. These patients also had higher levels of creatinine (and thus lower eGFR), cholesterol, triglycerides, uric acid and ESR. There were no significant differences between the two groups concerning other demographic, clinical, radiographic, laboratory and active treatment data at baseline.
Long-term renal outcomes of the new CKD cases
Doubling of baseline creatinine occurred in 18 (18.3%) patients at a median of 12.6 (4.6, 18.5) years from clinic entry. 49 (50%) patients developed a sustained ≥40% reduction in baseline eGFR at a median of 12.2 (7.2, 18.1) years. Only two patients developed an eGFR <15 mL/min/1.73 m2 and required renal replacement therapy.
Risk factors for CKD
We used time-dependent Cox regression models to identify the factors associated with the development of CKD. Patients who had CKD at baseline were not included in this analysis. In the univariate model (table 2) adjusted for age, sex and eGFR at baseline, factors significantly associated with the development of CKD included: BMI (HR 1.04, p=0.001), DM (HR 3.10, p<0.001), hypertension (HR 2.23, p=0.001), kidney stones (HR 2.07, p=0.01), IHD (HR 2.57, p=0.03), RDJC (HR 1.02, p=0.02), syndesmophytes and/or sacroiliitis (HR 1.64, p=0.03), triglycerides (HR 1.28, p<0.001), uric acid (1.16, p<0.001; per 50-unit increase in mmol/L) and daily use of NSAIDs (HR 1.71, p=0.02). In the multivariate Cox regression models adjusted for age, sex and eGFR at baseline (table 2), factors independently associated with the development of CKD included DM (HR 2.58, p<0.001), kidney stones (HR 2.14, p=0.01), RDJC (HR 1.02, p=0.02), uric acid (HR 1.21, p<0.001; 50-unit increase), daily use of NSAIDs (HR 1.77, p=0.02) and MTX use (HR 0.51, p=0.01).
Univariate and multivariate Cox regression analyses of risk factors for CKD
Discussion
Although CKD has been recognised as a comorbidity of psoriatic disease,2 10–12 there has been a paucity of studies investigating its occurrence in a PsA cohort. In our large longitudinal study, the prevalence of CKD in PsA was 9.2% (123 patients out of 1336). 98 (7.3%) patients developed CKD during follow-up at a median of 8.2 (2.8, 14.0) years. On long-term follow-up, 50% of the new patients with CKD developed significant eGFR decline (≥40% of the baseline value), mirroring the progressive nature of kidney disease in PsA. Only two patients developed ESKD on follow-up, however.
A cross-sectional study29 including 559 745 adults in primary care in 5 provinces across Canada (59.4% of which come from Ontario) from 2010 to 2015 found the prevalence of CKD to be 7.2%. The study, based on electronic medical records, might, however, have been biased by CKD detection in high-risk populations. Furthermore, the mean age of the patients was 48.5 (17.8) years, slightly higher than that in our cohort, with a significantly greater prevalence of DM and hypertension (19.6% and 33.6%, respectively), thus limiting the ability to perform direct comparisons. The prevalence of CKD in our cohort was lower than that found in other analyses in PsA (16%–18.8%).10 16 However, the number of patients in our study is significantly higher (1336 patients, as compared with 22910 and 7616), which would imply a greater reliability of the results.
In the multivariate Cox regression models, the following factors were independently associated with the development of CKD: DM, kidney stones, higher damaged joint count, higher uric acid levels, daily use of NSAIDs and use of MTX; the latter being protective. Higher BMI, hypertension, IHD, the presence of syndesmophytes and/or sacroiliitis and higher triglyceride levels were significant in the univariate, but not in the multivariate model.
The development of kidney disease is known to be closely linked to risk factors, including hypertension, DM, obesity, nephrolithiasis, dyslipidaemia and cardiovascular disease.30 31 These are all comorbidities known to be associated with PsA.32 33 Consequently, the emergence of any of these conditions can contribute to the onset of kidney disease. Given that each independently poses a risk for renal damage, the overall risk is further heightened when multiple factors coexist, as in PsA.
The higher damaged joint count reflects a greater disease burden, as it signifies previous ongoing inflammatory activity, thus the association with CKD is not surprising. Likewise, the correlation with NSAID use is not unexpected, while the Assessment of SpondyloArthritis International Society NSAID index34 could not be calculated in our analysis due to the historical nature of the database, daily use of NSAIDs, as opposed to non-use or use only as needed, was found to be a risk factor for CKD in PsA. NSAIDs are commonly used as initial therapy in PsA, for both peripheral and axial disease.35 The link between NSAIDs and CKD is well known, especially in the context of prolonged exposure and high cumulative doses.36–39 Apart from disrupting the compensatory vasodilatory effect of renal prostaglandins, NSAIDs can induce renal inflammatory responses via different mechanisms.40 The spectrum of NSAID-induced nephrotoxic effects is wide. This includes acute kidney injury, acute interstitial nephritis, acute or chronic papillary necrosis, nephrotic syndrome and progression to CKD, among others.36 Conditions such as hypertension and DM, commonly seen in PsA, further expose the patients to the adverse effects of NSAIDs.36 Moreover, NSAIDs themselves have been linked to hypertension,36 41 which is an established independent risk factor for the development of CKD.
While the use of MTX at higher doses in oncology has been linked to nephrotoxicity, there have been inconclusive results regarding the nephrotoxic effects of the low doses used in rheumatoid arthritis,42 43 which are generally similar to those used in PsA. In a secondary analysis for eGFR and kidney adverse events using the randomised double-blind, placebo-controlled Cardiovascular Inflammation Reduction Trial, low-dose MTX (≤25 mg per week) was found to be safe in patients with normal kidney function or mild-to-moderate CKD.44 In our cohort, the use of MTX was protective against the development of CKD. Interestingly, the protective effect was not observed with other systemic therapies, especially biologics. Despite limited evidence, MTX is widely used as an initial therapy in the treatment of psoriasis and PsA. In Ontario, MTX (and other conventional synthetic DMARDs) are required to be used as first-line therapies before access to biological or targeted synthetic DMARDs in PsA. A possible anti-inflammatory effect, exerted at an early stage of disease-onset (and thus not seen with advanced therapies), may be related to the renoprotective effect. This harmonises with some evidence that suggests that the introduction of MTX in early-onset PsA may offer reasonable efficacy and drug retention.45 46
Our study has several strengths, including the large number of patients, the organised and high-quality data, the longitudinal design and the long-term follow-up. We acknowledge some limitations, including the retrospective nature of the study and the absence of data on proteinuria, renal imaging and biopsies, which play a role in the workup of kidney disease. The lack of a matched non-PsA cohort for data comparison is another limitation.
In conclusion, CKD is not infrequent in patients with PsA. Its development is associated with related comorbidities, joint damage and the use of NSAIDs. MTX appears to provide a renoprotective effect. Better control of risk factors and proper use of medications, besides having other health-related benefits, will also help preserve renal function.
Data availability statement
Data are available on reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
This research study was approved by the University Health Network Research Ethics Board (REB Study#: 08-0630). All individuals have provided written informed consent for this research study.
References
Footnotes
X @dafnaGladman
Presented at This study was presented as a poster at the EULAR 2023 meeting. https://ard.bmj.com/content/83/Suppl_1/713.2.
Contributors All authors were involved in the study conception and design, analysis, and interpretation of data. All authors approved the final version of the article to be published. Study conception and design: FK, SG, SA-M, RJC, VC and DG. Acquisition of data: VC and DG. Analysis and interpretation of data: FK, SG, SA-M, RJC, VC and DG. DG is the guarantor.
Funding The Gladman-Krembil Psoriatic Arthritis Research Program is supported by a grant from the Krembil Foundation.
Competing interests None declared.
Patient and public involvement statement Patients were not involved in the design of this research project.
Provenance and peer review Not commissioned; externally peer reviewed.