Arthritis & Rheumatism (Arthritis Care & Research) Vol. 49, No. 5, October 15, 2003, pp 735–740 © 2003, American College of Rheumatology LETTERS DOI 10.1002/art.11376 New standards for uric acid excretion and evidence for an inducible transporter To the Editor: In a recent article, Perez-Ruiz et al summarize their studies of uric acid excretion in 100 men with gouty arthritis and in 72 age- and sex-matched controls (1). Two aspects of this important work seem particularly noteworthy. The 24-hour excretion in gouty subjects (628 ⫾ 154 mg/day/1.73 m2) covers essentially the same range as that of the matched controls (594 ⫾ 143 mg/day/1.73 m2). Despite these apparently comparable ﬁndings, 35 of the gouty individuals were considered to be overexcreters, i.e., their excretion rate was considered to be abnormally high. My concern is that the cut-off point of 700 mg/day/1.73 m2 that was chosen here (and is widely used elsewhere) lies well within 1 standard deviation of the normal mean and therefore lacks both statistical and physiologic validity. It would be helpful to see the individual values from normal and gouty subjects plotted in parallel vertical arrays to permit a visual comparison of the frequency distributions. I suspect that it then would become clear that a more appropriate cut-off of 880 mg/day/1.73 m2 (2 standard deviations above the normal mean) would avoid categorizing as overexcretors most of the individuals so classiﬁed in this report. The control group reported by Perez-Ruiz et al is large, current, normalized to a constant body size, and well studied. It seems appropriate to use this population to reset the normal upper limit of daily uric acid excretion at 880 mg/1.73 m2 (while bearing in mind that the precision of this test is poor and no overexcretor should be so labeled on the basis of any single high value) (2). The clearance of uric acid (Cu; in milliliter minute/1.73 meter2) (as well as its corollary the fractional excretion [FEu; in percent]) remained remarkably constant in 58 gouty patients studied before and after their hyperuricemia was controlled by allopurinol. This ﬁnding stands in marked contrast to those of previous studies of uric acid excretion in individuals studied at baseline, after allopurinol administration, or after hyperuricemia was induced by an oral bolus of yeast RNA (metaanalyzed in reference 3). Those studies consistently found the urinary excretion of uric acid to vary as an exponential function of the serum level, i.e., the efﬁciency of excretion increased when the serum level rose and fell when the serum did too. Is it possible to reconcile these disparate ﬁndings? I think maybe so. A plausible hypothesis suggests that the transporter responsible for tubular reabsorption may be inducible by its substrate. Thus, more ﬁltered urate leads to more carrier, whereas a lower burden results in less carrier. This idea is teleologically attractive (why should the tubule be bur- dened with more carriers than it needs?), it has abundant exemplars in the realm of enzymology, and it has been demonstrated in other transcellular transport systems (including some in the renal tubules). If this feedback system involves a signiﬁcant response time, an acute urate load may overwhelm the tubular reabsorptive capacity, while an acute fall may lead to supereffective reabsorption as an excess of transporters competes for both ﬁltered and secreted urate (it is thought that absorption and secretion may be coextensive systems within the proximal tubule) (4). Additional support for the hypothesis may lie in the remarkably high clearance of infused uric acid (5), the marked overexcretion seen in acute leukemia and the tumor lysis syndrome (which may occur with only moderate hyperurcemia) (6), and (on the other end of the scale) the sometimes sluggish response of the serum urate level after institution of allopurinol therapy. Now that the apparent transporter has been identiﬁed and sequenced, the hypothesis of inducibility should be testable on a molecular level (7). The relevance of this hypothesis to the earlier exponential evidence lies in the fact that past studies tended to be short term, whereas I suspect that a longer time may have elapsed between the pre- and posttherapy observations of Perez-Ruiz et al. Up- or downregulation of the reabsorptive capacity may require a substantial period after a change in the ﬁltered load (days or weeks) before balance is restored at the previous clearance rate. Clearly, the hypothesis also implies that inefﬁcient excretion (with resultant hyperuricemia) may be caused not by defective carriers, but by a lower setting of the undiscovered feedback system that regulates their number. Do these new data mean that we should change the way we evaluate the renal component of our patients’ hyperuricemia? I think that the best answer is yes, but not by much. It remains desirable to identify the unusual patient who signiﬁcantly overexcretes urinary uric acid, and I believe the best way to do so is to screen with the Eu/GF that is easily measured in midmorning spot urine samples by taking the product of the urinary uric acid and serum creatinine and dividing by the urinary creatinine. The result is the milligrams of urinary uric acid per deciliter of glomerular ﬁltrate or, alternatively, it may be considered the excretion rate of uric acid, in milligrams/minute, after the glomerular ﬁltration rate (GFR) has been normalized to 100 ml/minute. The test is simple, it is uncomplicated by 24-hour collection problems, and it includes (in the GFR) a meaningful correction for differences in body size. If 2 or more such determinations are signiﬁcantly elevated (greater than 0.6 mg/dl), then more extensive evaluations may be made starting with 3 24-hour collections on a low purine diet (2). The quantitative information obtained by the Eu/GF may be extended to the qualitative realm by simply dividing it 735 736 by the serum urate concentration. This yields the FEu that Perez-Ruiz et al found to be unchanged when it was examined before and after allopurinol therapy. This value is the decimal fraction of uric acid in the glomerular ﬁltrate that is ultimately excreted in the urine. Although the bidirectional tubular transport of uric acid means that virtually all ﬁltered urate is reabsorbed and most of the excreted molecules enter the nephron by tubular secretion, the FEu remains a useful measure of overall excretion efﬁciency. Again using approximately 2 standard deviations from the mean of the new Spanish data, inefﬁcient excretion begins when the FEu falls below 4%. It must be noted here that the utility of clearance (or FEu) has long been emphasized by other observers, most notably Brian Emmerson (8). As a ﬁnal thought, may I suggest that we try to do away with the term underexcretion? Generations of medical students have been confused by the idea that overexcretion of uric acid is associated with hyperuricemia and gout, and, by the way, so is underexcretion. The fact that the ﬁrst term is quantitative but the second is qualitative often escapes the naı̈ve reader. To me, it makes sense to retain overexcretion for those unusual patients who overproduce urate with a resultant abnormally high excretion rate of uric acid. For those who cope badly with a normal urate burden, “inefﬁcient excretion” seems a more clear and useful descriptor. Doesn’t it make sense to say simply that hyperuricemia results from making too much uric acid (which we can evaluate by studying the excretion rate) and/or from clearing it inefﬁciently (which we evaluate by measuring the fractional excretion)? Peter A. Simkin, MD University of Washington, Seattle 1. Perez-Ruiz F, Calabozo M, Garcı́a Erauskin G, Ruibal A, Herrero-Beites AM. Renal underexcretion of uric acid is present in patients with apparent high urinary uric acid output. Arthritis Rheum (Arthritis Care Res) 2002;47:610 –3. 2. Simkin PA. When, why, and how should we quantify the excretion rate of urinary uric acid? J Rheumatol 2001;28:1306 – 10. 3. Simkin PA. Uric acid excretion in patients with gout. Arthritis Rheum 1979;22:98 –9. 4. Steele TH. Hyperuricemic nephropathies. Nephron 1999;81: 45–9. 5. Berliner RW, Hilton JG, Yü TF, Kennedy TJ. The renal mechanism for urate excretion in man. J Clin Invest 1950;29:396 – 401. 6. Rieselbach RE, Bentzel CJ, Cotlove E, Frei E, Freireich EJ. Uric acid excretion and renal function in the acute hyperuricemia of leukemia. Am J Med 1964;37:872– 84. 7. Enomoto A, Kimura H, Chairoungdua A, Shigeta Y, Jutabha P, Cha SH, et al. Molecular identiﬁcation of a renal urate anion exchanger that regulates blood urate levels. Nature 2002;417: 447–52. 8. Emmerson BT. The management of gout. N Engl J Med 1996; 334:445–51. DOI 10.1002/art.11373 Reply To the Editor: We thank Dr. Peter Simkin’s for his comments. Nevertheless, some points should be emphasized. Letters Figure 1. Correlation between 24-hour urinary uric acid (Uur) and Simkin’s Index stratiﬁed by creatinine clearance (Ccr). GF ⫽ glomerular ﬁltration. First, we used 700 mg/day/1.73 m2 as the cut-off value for 24-hour urinary uric acid because it is widely used in clinical practice and only to ascertain if it was an appropriate cut-off point for classifying patients with gout. I agree with Dr. Simkin that 880 mg/day/1.73 m2 may be a more appropriate value. Indeed, we designed the study because we thought that 700 mg/day/1.73 m2 was not the proper cut-off point. Second, the fact that the clearance of uric acid did not change signiﬁcantly in patients taking allopurinol may be interpreted as follows. Our patients did not have exogenously induced hyperuricemia or hypouricemia after high doses of allopurinol, and data were strictly paired, contrary to other studies (1). Thus, the levels of serum uric acid during therapy were within those observed in clinical practice during urate-lowering therapy. Indeed, we have recently published a retrospective study on the long-term efﬁcacy of urate-lowering therapy in a large cohort of renal transplant patients. Again, clearance of urate was unchanged in patients on allopurinol therapy (2). Dr. Simkin’s hypothesis suggesting that the tubular transporter for reabsorption may be an inducible one is attractive and surely warrants further investigation to ascertain if this is true in any range of serum urate level. Third, I also agree with Dr. Simkin about the usefulness of the excretion of uric acid per volume of glomerular ﬁltration (EuGF) test in everyday clinical practice— easy to do, cheap, and showing a good correlation with other parameters— but related to renal function (3–5). In fact, since the serum creatinine to urinary creatinine ratio (Pcr/ Ucr) quotient may rise in patients with signiﬁcant renal function impairment, this increase may result in a high EuGF result. Thus, patients showing moderate to severe renal function impairment may show EuGF ⬎ 0.6 mg/dl and may be misclassiﬁed as overproducers. They are not actually overproducers, but relative good excretors when considering the low glomerular ﬁltration rate, as it occurs with fractional excretion of urate in patients with signiﬁcant renal function impairment. Figure 1 shows the good correlation between 24-hour urinary uric acid and EuGF Letters when patients are stratiﬁed according to clearance of creatinine in our cohort of 404 gouty patients (unpublished data). Nearly 50% of the patients with apparent overproduction according to EuGF due to renal insufﬁciency will not show serum creatinine values ⬎1.5 mg/dl (most of them older patients with low body mass index). Thus, patients showing EuGF ⬎ 0.6 mg/dl, even with “normal” serum creatinine, should be tested with 24-hour urinary samples, and clearance of urate and creatinine should be calculated. Finally, the suggestion of eliminating the term “underexcretion” is certainly plausible, but “overexcretion” should also be eliminated. Indeed, patients with overproduction and normal renal function will show apparent overexcretion if compared at hyperuricemia with normal controls at normal serum urate levels (6), but similar uric acid excretion when both groups show similar glomerular ﬁltered load of urate (6). Thus, we suggest that “inefﬁcient excretion” and “efﬁcient excretion” may be more proper physiopathologic terms than “underexcretion” or “overexcretion,” respectively, when we are talking about renal handling of uric acid. In addition, we suggest that EuGF should be changed to Simkin’s Index as a tribute to Dr. Simkin’s work in this, most of the times undervalued, ﬁeld of gout. Fernando Perez-Ruiz, MD Ana Herrero-Beites, MD Hospital de Cruces, Baracaldo, Vizcaya, Spain 1. Puig JG, Mateos F, Lopez M, Conthe P. Renal handling of uric acid in gout: impaired tubular transport of urate not dependent on serum urate levels. Metabolism 1986;35:1147–53. 2. Perez-Ruiz F, Gomez-Ullate P, Amenabar JJ, Zarraga S, Calabozo M, Herrero-Beites A. Long-term efﬁcacy of urate-lowering drugs in renal transplant patients. Nephrol Dial Trasplant 2003;18:603– 6. 3. Perez-Ruiz F, Calabozo M, Alonso-Ruiz A, Ruiz-Lucea E, Herrero-Beites A. Analysis of the methods for the classiﬁcation of renal handling of urate in gout. Arthritis Rheum 1997;40(9 suppl):S48. 4. Perez Ruiz F, Calabozo Raluy M, Herrero Beites A, Ruiz Lucea E, Alonso Ruiz A. Analysis of the methods for classifying gout according to renal excretion of uric acid. Rev Esp Reumatol 1998;25:335–9. 5. Perez-Ruiz F, Calabozo M, Herrero-Beites AM, Ruibal A, Alonso-Ruiz A. Inﬂuence of serum uric acid levels and glomerular ﬁltration rate on the tests used to classify patients with gout. Arthritis Rheum 2001;44(9 suppl):S128. 6. Perez-Ruiz F, Calabozo M, Garcia-Erauskin G, Ruibal A, Herrero-Beites AM. Renal underexcretion of uric acid is present in patients with apparent high urinary uric acid output. Arthritis Rheum 2002;47:610 –3. DOI 10.1002/art.11371 Hodgkin’s lymphoma presenting as Tietze’s syndrome To the Editor: We read with great interest the report of Fioravanti et al about malignant lymphoma presenting as Tietze’s syndrome (Fioravanti A, Toﬁ C, Volterrani L, Marcolongo R. Malignant lymphoma presenting as Tietze’s syndrome. Ar- 737 thritis Rheum [Arthritis Care Res] 2002;47:229 –30). We would like to add to the 4 reported cases a patient seen recently at our medical center. A 26-year-old woman was seen at the American University of Beirut Medical Center with pain over the upper parasternal area of several months duration. The pain was worse at night and associated with morning stiffness in the chest for 30 minutes. No other associated symptoms were reported. Initial blood studies, including a complete blood count, erythrocyte sedimentation rate, rheumatoid factor, antinuclear antibodies, and chemistry proﬁle, yielded results within normal guidelines. Her pain responded very well to nonsteroidal antiinﬂammatory drugs; however, each time these drugs were discontinued, her pain would recur. Six months after her initial presentation, a radiograph of the chest and costochondral joints did not reveal any abnormalities. The patient presented a few months later with a bulging mass over the upper chest wall. A chest computed tomography scan showed an anterior mediastinal mass measuring 11 cm ⫻ 9.5 cm. Osteolytic involvement of the whole sternum and proximal adjacent aspect of the ribs was also noted. The lung parenchyma was clear. An excisional biopsy revealed a Hodgkin’s lymphoma of the nodular sclerosis type. Our case adds additional support to the view that malignancy should always be suspected in the differential diagnosis of Tietze’s syndrome, and that the use of the appropriate radiographic studies will help in the early detection of these tumors. Imad Uthman, MD Ihab El-Hajj, MD Rana Traboulsi, MD Ali Taher, MD American University of Beirut, Beirut, Lebanon DOI 10.1002/art.11372 Neovascularization around the optic nerve in giant cell arteritis To the Editor: Neovascularization in giant cell arteritis (GCA) has been described (1). The response of the artery to the injury includes mobilization and proliferation of smooth muscle cells in conjunction with matrix production and angiogenesis (2). Increased tissue and serum angiogenic activity has been associated with low prevalence of ischemic complications in patients with GCA (3). We report a GCA patient with several episodes of amaurosis fugax. Magnetic resonance imaging (MRI) disclosed collateral circulation around the left optic nerve. A 65-year-old man was sent to our hospital because of episodes of amaurosis fugax of 6 months duration. One year prior, he had been diagnosed with biopsy-proven GCA in another center. After diagnosis, he was started on treatment with prednisone. However, within the 6 months prior to admission he complained of left supraorbitary headache and suffered several episodes of left amaurosis 738 Letters Carlos Garcia-Porrua, MD, PhD Robustiano Pego-Reigosa, MD, PhD Victor Armesto, MD Miguel A. Gonzalez-Gay, MD, PhD Hospital Xeral-Calde, Lugo, Spain 1. Kaiser M, Younge B, Bjornsson J, Goronzy JJ, Weyand CM. Formation of new vasa vasorum in vasculitis: production of angiogenic cytokines by multinucleated giant cells. Am J Pathol 1999;155:765–74. 2. Weyand CM, Goronzy JJ. Pathogenic principles in giant cell arteritis. Int J Cardiol 2000; 31;75 Suppl 1:S9 –15; discussion S17–9. 3. Cid MC, Hernandez-Rodriguez J, Esteban MJ, Cebrian M, Gho YS, Font C, et al. Tissue and serum angiogenic activity is associated with low prevalence of ischemic complications in patients with giant-cell arteritis. Circulation 2002;24:1664 –71. 4. Cid MC, Font C, Oristrell J, de la Sierra A, Coll-Vinent B, Lopez-Soto A, et al. Association between strong inﬂammatory response and low risk of developing visual loss and other cranial ischemic complications in giant cell (temporal) arteritis. Arthritis Rheum 1998;41:26 –32. Figure 1. Orbital magnetic resonance image showing collateral circulation around the left optic nerve. DOI 10.1002/art.11374 Education for patients with rheumatoid arthritis: is it worth it? fugax. No adequate information was obtained about the initial dosage of prednisone and the way in which prednisone dosage was tapered. Despite having episodes of amaurosis fugax, he was receiving only 10 mg of prednisone per day at the time of admission. Physical examination including funduscopy did not disclose abnormalities. Laboratory data, including coagulation test results, anticardiolipin antibodies, full blood cell count, and hepatic and renal function parameters, were negative or normal. Erythrocyte sedimentation rate was 41 mm/ hour (normal ⬍20 mm/hour) and C-reactive protein was 12 mg/liter (normal ⬍5 mg/liter). Plain chest radiograph and carotid angio-MRI did not show abnormalities. Orbital and cerebral MRI showed collateral circulation around the left optic nerve (Figure 1). Treatment with acetylsalicylic acid (300 mg/day) was started and prednisone dosage was increased to 40 mg/ day. Following this procedure, a dramatic improvement of headache was achieved. Also, no new visual complications have occurred over the following 6 months. Inﬂammation of the artery wall in GCA induces a series of structural changes, including the formation of new vasa vasorum. Angiogenesis has been proven more prominent in patients with a strong acute phase response (4). In our patient, an inappropriately low dose of prednisone might have been responsible for the persistent inﬂammatory response, manifested by headache and elevation of the inﬂammatory laboratory markers. It is possible that the persistence of chronic inﬂammatory response might have played a role in the striking neovascularization phenomena around the optic nerve. Based on that, we are tempted to speculate that neovascularization may be a compensatory mechanism to maintain an adequate blood supply to the optic nerve in GCA. To the Editor: Rheumatoid arthritis (RA) is the most frequent chronic inﬂammatory disease affecting approximately 1% of the white population, particularly females (3 times more often than males) (1). Because of its articular and extraarticular manifestations and complications, such as infections and osteoporosis, RA has a considerable impact on a patient’s quality of life, with mayor physical, psychological, and social consequences. Treatment of RA includes not only medications but also education and cognitive-behavioral interventions (2). Recently, Riemsma et al described the impact of the use of education as a therapeutic tool in patients with RA (3). These authors proposed a categorization of educational practices into information only, counseling, and behavioral therapy. This approach is important because it emphasizes the various types of educational systems and allows analysis of data more objectively to avoid comparisons of unrelated interventions. Surprisingly, Riemsma’s results indicated only a modest and short-term effect of education on RA patients, including global assessment, psychological status, and general well being (3). However, education may not only be important for the psychological well being of the patients but also for improving their adherence to therapy and for guiding patient’s participation on therapeutic decisions. Indeed, adherence tends to be higher in patients who have participated in educational programs (4,5). Moreover, the trend toward a favorable effect of education on depression may warrant its use as a therapeutic tool in patients with RA due to its known association with adherence (6). The clinical expression of RA varies among populations and is inﬂuenced by several factors, including race, genetics, environmental factors, and access to medical care (7). Letters 739 Table 1. Level of knowledge about rheumatoid arthritis (RA) in patients compared with the general population (no RA) Characteristic Sex, male:female Age, years, mean ⫾ SD Education, % None Primary school High school Technical University Postgraduate Socioeconomic status, %* 2 3 4 Correct answers to true or false questions, % 1. There is only a single therapy for all RA patients (Existe un tratamiento único para los pacientes con artritis reumatoidea) 2. All RA patients have a bad prognosis (Todos los pacientes con artritis reumatoidea tienen un mal pronóstico) 3. The neck is the most commonly affected area of the spine in patients with RA (El cuello es la parte de la columna que se afecta mas comúnmente en los pacientes con artritis reumatoidea) 4. RA is caused by cold weather, inadequate nourishment and humidity (La artritis reumatoidea es causada por la mala alimentación, el frı́o y el clima húmedo) 5. Patients with RA should not play a role in the management of their disease because the physician is the one in charge (Los pacientes con artritis reumatoidea no deben asumir un papel muy activo en su proceso de recuperación porque el médico es el encargado principal) 6. RA and osteoporosis are the same disease (Artritis y osteoporosis son la misma enfermedad) 7. RA can produce dryness in the eyes and mouth (La artritis reumatoidea puede causar resequedad en los ojos y en la boca) 8. Patients with RA should include all kinds of food in their meals but must avoid meat (Los pacientes con artritis reumatoidea deben incluir en su dieta alimentos variados pero no deben comer carnes rojas) 9. RA never compromises the lungs (La artritis reumatoidea nunca afecta los pulmones) 10. To conﬁrm the diagnosis of RA, it is necessary to perform some blood tests (Para conﬁrmar el diagnóstico de artritis reumatoidea es necesario hacer algunos exámenes de sangre) More than 6 correct answers, % Number of correct answers, mean ⫾ SD RA (n ⴝ 448) No RA (n ⴝ 269) 44:402 54 ⫾ 13 40:227 49.3 ⫾ 16 2 38 35 13 10 2 1 26 32 14 22 5 29 44 14 25 44 16 58 68† 43 58† 38 27 56 59 59 64 80 85 52 33† 33 30 26 31 87 75† 47 4.97 ⫾ 2.31 48 5 ⫾ 2.36 * According to the Colombian socioeconomic classiﬁcation. † P ⬍ 0.01 after adjusting for age and education level. Disability is one of the most important consequences of RA and is not free from the inﬂuence of sociocultural factors, such as attitudes towards health, illness, the patient’s sociocultural level (8). Physicians should be aware of these factors before extrapolating ﬁndings on RA populations different from their own. In addition, RA treatment and outcome are better managed by rheumatologists than by other physicians (9). The state of Antioquia (capital, Medellı́n) is geographically located in northwestern Colombia between the central and western branches of the Andean Mountains and has a population of 4,500,000. A large number of patients in this area do not know which physician should manage their illness, and access to specialized rheumatology care is limited by the restriction of social security laws, as well as by the low number (only 11) of rheumatologists in practice. A sizeable number of patients are treated with alternative medicine while others have no treatment at all. We wanted to explore the general knowledge about RA of patients attending our rheumatology unit, as well as that of the general population. To that purpose, we surveyed 448 RA patients and 269 healthy individuals, using a 10-point questionnaire on RA knowledge (10) (Table 1). The questionnaire had been validated with a pilot study and through construct validity analyses (11). Patients with RA knew more about the disease’s symptoms than healthy persons but the latter had more knowledge concerning prognosis and therapy than the patients. There were no 740 Letters signiﬁcant differences in knowledge about the etiology of disease in these 2 groups. The poor knowledge about RA in both patients and the general population, and the ﬁnding that patients had the same level of knowledge as the general population, suggest that educational programs are sorely needed to provide adequate and rational information on RA. This opens the discussion concerning the need to balance the importance of education as a therapeutic tool, and the ethical and social implications of providing education in an environment where managed care and cost-beneﬁt issues are the prevailing forces. Jose Cadena, MD Juan-Manuel Anaya, MD Universidad Pontiﬁcia Bolivariana, Medellı́n, Colombia 4. 5. 6. 7. 8. 9. 1. Lawrence RC, Helmick CG, Arnett FC, Deyo RA, Felson DT, Giannini EH, et al. Estimates of the prevalence of arthritis and selected musculoskeletal disorders in the United States. Arthritis Rheum 1998;41:778 –99. 2. American College of Rheumatology Subcommittee on Rheumatoid Arthritis Guidelines. Guidelines for the management of rheumatoid arthritis, 2002 update. Arthritis Rheum 2002; 46:328 – 46. 3. Riemsma RP, Kirwan JR, Taal E, Rasker JJ. Patient education 10. 11. for adults with rheumatoid arthitis. Cochrane Database Syst Rev 2002;3:CD003688. Hill J, Bird H, Johnson S. Effect of patient education on adherence to drug treatment for rheumatoid arthritis: a randomised controlled trial. Ann Rheum Dis 2001;60:869 –75. Viller F, Guillemin F, Briancon S, Moum T, Suurmeijer T, van den Heuvel W. Compliance to drug treatment of patients with rheumatoid arthritis: a 3 year longitudinal study. J Rheumatol 1999;26:2114 –22. DiMatteo MR, Lepper HS, Croghan TW. Depression is a risk factor for noncompliance with medical treatment: meta-analysis of the effects of anxiety and depression on patient adherence. Arch Intern Med 2000;160:2101–7. Anaya JM, Correa PA, Mantilla RD, Jimenez F, Kuffner T, McNicholl JM. Rheumatoid artritis in African Colombians from Quibdo. Semin Arthritis Rheum 2001;31:191– 8. Escalante A, del Rincón I. The disablement process in rheumatoid arthritis. Arthritis Rheum 2002;47:333– 42. Emery P, Breedveld FC, Dougados M, Kalden JR, Schiff MH, Smolen JS. Early referral recommendation for newly diagnosed rheumatoid arthritis: evidence based development of a clinical guide. Ann Rheum Dis 2002;61:290 –7. Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, et al. The American Rheumatism Association 1987 revised criteria for the classiﬁcation of rheumatoid arthritis. Arthritis Rheum 1988;31:315–24. Cadena J, Alvarez A, Correa M, Bonilla LM, Gómez MP, Montoya MA, et al. Encuesta de conocimiento sobre artritis reumatoidea: ¿vale la pena educar? Rev Colomb Reumatol 2002;9:262–9. DOI 10.1002/art.11395 Erratum In the article by McQuillan et al published in the June 2003 issue of Arthritis Care & Research (pp 368 –76), the headings False Positive and True Positive on Figure 2 were transposed, and the last column heading on Table 4 should read 1⫺Speciﬁcity. Also, in the ﬁrst paragraph of the Results section it should read “current affective disorder” instead of “current major depressive episode,” and the last sentence in the Results should read “. . . we found that the sensitivity (0.89) is much better than 1⫺ the speciﬁcity (0.24) at the standard cut off score of 16. We regret the errors.