Anti-tumour necrosis factor therapy in rheumatoid arthritis and risk of malignant lymphomas: relative risks and time trends in the Swedish Biologics Register

BACKGROUND

Tumour necrosis factor (TNF) antagonists have proved effective as treatment against rheumatoid arthritis (RA), but the unresolved issue of whether the use of anti-TNF therapy increases the already elevated risk of lymphoma in RA remains a concern.

METHODS

Using the Swedish Biologics Register (ARTIS), the Swedish Cancer Register, pre-existing RA cohorts and cross-linkage with other national health and census registers, a national RA cohort (n = 67,743) was assembled and patients who started anti-TNF therapy between 1998 and July 2006 (n = 6604) were identified. A general population comparator (n = 471,024) was also assembled and the incidence of lymphomas from 1999 to 31 December 2006 was assessed and compared in these individuals.

RESULTS

Among the 6604 anti-TNF-treated RA patients, 26 malignant lymphomas were observed during 26,981 person-years of follow-up, which corresponded to a relative risk (RR) of 1.35 (95% CI 0.82 to 2.11) versus anti-TNF-naive RA patients (336 lymphomas during 365,026 person-years) and 2.72 (95% CI 1.82 to 4.08) versus the general population comparator (1568 lymphomas during 3,355,849 person-years). RA patients starting anti-TNF therapy in 1998-2001 accounted for the entire increase in lymphoma risk versus the two comparators. By contrast, RR did not vary significantly by time since start of first treatment or with the accumulated duration of treatment, nor with the type of anti-TNF agent.

CONCLUSION

Overall and as used in routine care against RA, TNF antagonists are not associated with any major further increase in the already elevated lymphoma occurrence in RA. Changes in the selection of patients for treatment may influence the observed risk.

Haematopoietic malignancies in rheumatoid arthritis: lymphoma risk and characteristics after exposure to tumour necrosis factor antagonists

BACKGROUND

Patients with rheumatoid arthritis (RA) are at increased risk of malignant lymphomas, and maybe also of leukaemia and multiple myeloma. The effect of tumour necrosis factor (TNF) antagonists on lymphoma risk and characteristics is unclear.

OBJECTIVE

To assess expected rates and relative risks of haematopoietic malignancies, especially those associated with TNF antagonists, in large population based cohorts of patients with RA.

METHODS

A population based cohort study was performed of patients with RA (one prevalent cohort (n = 53,067), one incident cohort (n = 3703), and one TNF antagonist treated cohort 1999 through 2003 (n = 4160)), who were linked with the Swedish Cancer Register. Additionally, the lymphoma specimens for the 12 lymphomas occurring in patients with RA exposed to TNF antagonists in Sweden 1999 through 2004 were reviewed.

RESULTS

Study of almost 500 observed haematopoietic malignancies showed that prevalent and incident patients with RA were at increased risk of lymphoma (SIR = 1.9 and 2.0, respectively) and leukaemia (SIR = 2.1 and 2.2, respectively) but not of myeloma. Patients with RA treated with TNF antagonists had a tripled lymphoma risk (SIR = 2.9) compared with the general population. After adjustment for sex, age, and disease duration, the lymphoma risk after exposure to TNF antagonists was no higher than in the other RA cohorts. Lymphomas associated with TNF antagonists had characteristics similar to those of other RA lymphomas.

CONCLUSION

Overall, patients with RA are at equally increased risks for lymphomas and leukaemias. Patients with RA treated with TNF antagonists did not have higher lymphoma risks than other patients with RA. Prolonged observation is needed to determine the long term effects of TNF antagonists on lymphoma risk.

Risks of solid cancers in patients with rheumatoid arthritis and after treatment with tumour necrosis factor antagonists

BACKGROUND

Existing studies of solid cancers in rheumatoid arthritis (RA) reflect cancer morbidity up until the early 1990s in prevalent cohorts admitted to hospital during the 1980s.

OBJECTIVE

To depict the cancer pattern of contemporary patients with RA, from updated risk data from prevalent and incident RA populations. To understand the risk of solid cancer after tumour necrosis factor (TNF) treatment by obtaining cancer data from cohorts treated in routine care rather than trials.

METHODS

A population based study of three RA cohorts (one prevalent, admitted to hospital 1990-2003 (n = 53,067), one incident, diagnosed 1995-2003 (n = 3703), and one treated with TNF antagonists 1999-2003 (n = 4160)), which were linked with Swedish nationwide cancer and census registers and followed up for cancer occurrence through 2003.

RESULTS

With 3379 observed cancers, the prevalent RA cohort was at marginally increased overall risk of solid cancer, with 20-50% increased risks for smoke related cancers and +70% increased risk for non-melanoma skin cancer, but decreased risk for breast (-20%) and colorectal cancer (-25%). With 138 cancers, the incident RA cohort displayed a similar cancer pattern apart from non-decreased risks for colorectal cancer. TNF antagonist treated patients displayed solid cancer (n = 67) risks largely similar to those of other patients with RA.

CONCLUSION

The cancer pattern in patients treated with TNF antagonists mirrors those of other contemporary as well as historic RA cohorts. The consistent increase in smoking associated cancers in patients with RA emphasises the potential for smoking cessation as a cancer preventive measure in RA.

Results from a nationwide postmarketing cohort study of patients in Sweden treated with etanercept

OBJECTIVES

To describe a nationwide system for postmarketing follow up of new antirheumatic drugs in Sweden, and to analyse safety and effectiveness in an etanercept treated patient cohort.

METHODS

Etanercept became available in Sweden for prescribing on a named patient basis in 1999. All patients treated were included in a follow up of intensified adverse event reporting and recording of clinical outcome during 24 months, according to the EULAR core set.

RESULTS

The mean (SD) disease activity score (DAS 28) value at inclusion among 820 patients recruited on a named patient basis during year 1 was 5.99 (1.19). After two years, 21% (n = 172) of these patients had discontinued the treatment. Of the remaining 648 patients, 68% (n = 442) responded to the treatment. However, in 55% of the responders, the disease activity was intermediate or high (mean DAS 28, 3.37 (1.20)). In all, 540 adverse events were reported in 421 adverse drug reaction (ADR) reports, in 294 patients. The events in 80 reports (19%) were serious. Twenty two per cent of the events were infections, of which 24% (n = 29) were serious. The incidence of serious adverse events remained constant over time.

CONCLUSIONS

At start of etanercept treatment, patients had high disease activity. Activity remained high in a large proportion of the responding patients. Although serious ADRs occurred during late phases of treatment, no unexpected safety problems arose. No specific indicators of ADR risk were found. The monitoring system that was established may be useful in future postmarketing surveillance.

Increased endothelial expression of HLA-DQ and interleukin 1alpha in extra-articular rheumatoid arthritis. Results from immunohistochemical studies of skeletal muscle

OBJECTIVE

To investigate markers of endothelial activation in muscle biopsies from rheumatoid arthritis (RA) patients with and without extra-articular manifestations (ExRA).

PATIENTS AND METHODS

Nine consecutive ExRA patients were compared with nine RA controls without ExRA, matched for age, sex and duration of RA. Muscle biopsies were obtained from the lateral vastus or anterior tibial muscle. Macrophage and lymphocyte CD markers, HLA molecules, cytokines and adhesion molecules were investigated using immunohistochemistry, and stainings were evaluated using computer image analysis and conventional microscopy. Serum concentrations of soluble adhesion molecules, tumour necrosis factor alpha (TNF-alpha) and rheumatoid factor (RF) were determined using immunoassays.

RESULTS

The number of HLA-DQ-positive capillaries (P=0.039) and the expression of interleukin 1alpha (IL-1alpha) in endothelial cells (mean pairwise difference 0.26%; 95% confidence interval 0-0.52) were increased in ExRA patients compared with non-ExRA controls. There were no signs of inflammatory cell infiltrates or fibre degeneration. Serum levels of TNF-alpha, the soluble form of intercellular adhesion molecule 1, the soluble form of vascular cell adhesion molecule 1 and IgM RF were increased in the ExRA group.

CONCLUSION

The increased expression of HLA-DQ and IL-1alpha may indicate systemic endothelial activation in extra-articular RA, which could be of importance for cardiovascular comorbidity and mortality in such patients.

Seroprevalence of Helicobacter pylori in primary Sjögren's syndrome

OBJECTIVE

To study the seroprevalence of Helicobacter pylori (H. pylori) infection in patients with primary Sjögren's syndrome (SS), fulfilling the 1993 European classification criteria compared with three different control groups.

METHODS

Serological tests investigating the presence of antibodies against H. pylori were performed by Enzyme Immuno Assay (EIA) and confirmed by immunoblot (IB). The samples were tested for antibodies against cytotoxin-associated-protein A (CagA). The three control groups included were: one simultaneously collected age-matched group of orthopaedic outpatients without rheumatological disease, a random primary care patient sample from the same geographic region and a group of age-matched blood donors.

RESULTS

45% of the SS patients (n = 164) were EIA-positive for H. pylori and 30% were positive in the confirming IB assay. 23% had antibodies to the CagA protein. We found a clear and statistically significant increase in seroprevalence with increasing age. These estimates were lower compared to the control group of orthopaedic patients but similar to those in the other two control groups, thus showing the importance of multiple control groups in case control studies. In the group of SS patients there were no significant associations between a positive EIA, IB or CagA for H. pylori and the presence of abnormal serum levels of autoantibodies (ANA, anti-SSA, anti-SSB, rheumatoid factor (RF)) or an abnormal lip biopsy.

CONCLUSION

Swedish patients with primary SS do not have higher H. pylori seroprevalence rates than controls. Neither was H. pylori seropositivity associated with the presence of immunological markers of SS such as circulating autoantibodies or a lip biopsy with abnormal focus score.

Joint swelling as a predictor of death from cardiovascular disease in a population study of Pima Indians

OBJECTIVE

Markers of inflammation have recently been shown to be predictive of cardiovascular disease (CVD). Furthermore, the excess mortality in rheumatoid arthritis (RA), a disease characterized by chronic polyarthritis, is chiefly due to death from CVD. With this background, we studied the effect of inflammation, as reflected by the number of joints with soft tissue swelling, and rheumatoid factor (RF) seropositivity on CVD-related mortality.

METHODS

Mortality rates and rate ratios for all-cause and CVD-related deaths were computed in a longitudinal, population-based cohort of Pima Indians in Arizona from 1965 through 1994. Repeated health examinations were performed, involving systematic assessment of the features of RA, cardiovascular risk factors, serum titers of RF, as well as mortality. The cohort comprised 4,120 subjects (1,861 men, 2,259 women) who were examined an average of 3.5 times during a mean followup of 14 years.

RESULTS

During the followup period, 182 CVD-related deaths ocurred. The age- and sex-adjusted CVD-related mortality rates increased significantly with the presence of a higher number of joints with soft tissue swelling (Ptrend = 0.04), and were 2.07 (95% confidence interval [95% CI] 1.30-3.31) times as high in those subjects who had 2 or more swollen joints as in those who had none. There were no significant additional effects on CVD-related mortality when seropositivity for RF or a previous diagnosis of RA were considered. In age- and sex-adjusted proportional hazards analyses, which were controlled for possible confounders, the effect of swollen joints remained significant (mortality rate ratio 1.33, 95% CI 1.04-1.71 per category increase [no swollen joints, 1 swollen joint, at least 2 swollen joints]).

CONCLUSION

Joint swelling is a significant risk factor for CVD-related death, independent of other known risk factors including a diagnosis of RA. This finding supports the hypothesis that inflammatory mechanisms are important for the development of CVD.

Incidence of physician-diagnosed primary Sjögren syndrome in residents of Olmsted County, Minnesota

OBJECTIVES

To estimate the incidence of physician-diagnosed primary Sjögren syndrome (SS) among residents of Olmsted County, Minnesota, in the setting of usual medical care and to determine how often objective criteria are available in the medical records of such patients.

PATIENTS AND METHODS

We reviewed all medical records of residents in Olmsted County with physician-diagnosed SS from 1976 to 1992 to determine whether they had undergone objective tests for keratoconjunctivitis sicca, salivary dysfunction, or serologic abnormality. Confounding illnesses were excluded. To identify misclassified cases, all records from patients with xerostomia or keratoconjunctivitis sicca were also reviewed. The average annual SS incidence rates were calculated by considering the entire population to be at risk.

RESULTS

Of 75 patients with onset of SS during the study period, 53 had primary SS. All patients were white, 51 (96.2%) were women, and the mean +/- SD age was 59+/-15.8 years. The age- and sex-adjusted annual incidence was 3.9 per 100,000 population (95% confidence interval, 2.8-4.9) for patients with primary SS. Eleven patients (20.8%) with physician-diagnosed SS had no documentation of objective eye, mouth, or laboratory abnormalities. Objective evaluations performed most frequently were laboratory and ocular tests and least often were investigations of xerostomia.

CONCLUSIONS

The average annual incidence rate for physician-diagnosed primary SS in Olmsted County is about 4 cases per 100,000 population. These data probably underestimate the true incidence because they are based on usual medical care of patients with SS in a community setting, rather than on a case-detection survey. In the future, a true incidence may be possible with a higher index of suspicion, greater attention to objective tests, and increased awareness of new classification criteria for SS. For epidemiological studies based on existing data, application of current criteria may not be feasible, and consensus on criteria for such studies would be useful.

Chronic musculoskeletal pain, prevalence rates, and sociodemographic associations in a Swedish population study

OBJECTIVE

To estimate the prevalence of chronic regional and widespread musculoskeletal pain in a sample of the general adult population and study the association to age, sex, socioeconomic class, immigration, and housing area.

METHODS

A cross sectional survey with a postal questionnaire to 3928 inhabitants on the west coast of Sweden.

RESULTS

The age and sex adjusted prevalence of chronic regional pain (CRP) was 23.9% and chronic widespread pain (CWP) 11.4% among 2425 subjects who responded to the complete questionnaire. Odds ratio (OR) for CWP showed a systematic increasing gradient with age and was highest in the age group 59-74 yrs (OR 6.36, 95% CI 3.85-10.50) vs age group 20-34 yrs. CWP was also associated with female sex (OR 1.91, 95% CI 1.41-2.61), being an immigrant (OR 1.83, 95% CI 1.22-2.77), living in a socially compromised housing area (OR 3.05, 95% CI 1.48-6.27), and being an assistant nonmanual lower level employee (OR 1.92, 95% CI 1.09-3.38) or manual worker (OR 2.72, 95% CI 1.65-4.49) vs being an intermediate/higher nonmanual employee. OR for CRP showed a systematic increasing gradient with age and was highest in the age group 59-74 yrs (OR 2.22, 95% CI 1.62-3.05) vs age group 20-34 yrs. CRP was also associated with being a manual worker (OR 1.63, 95% CI 1.19-2.23) vs being an intermediate/higher nonmanual employee.

CONCLUSION

Chronic musculoskeletal pain is common in the general population. Sociodemographic variables were overall more frequently and strongly associated with CWP than with CRP, which indicates different pathophysiology in the development or preservation of pain in the 2 groups.

Sympathetic dysfunction in patients with primary Sjögren's syndrome

OBJECTIVE

To investigate autonomic nervous system function in patients with primary Sjögren's syndrome (SS) and relate the findings to clinical variables.

METHODS

Autonomic nervous system function was determined in 30 patients with primary SS using the finger skin blood flow test [vasoconstrictory (VAC) index], deep-breathing test [expiration/inspiration (E/I) ratio], and the tilt table (orthostatic) test [acceleration index (AI), brake index (BI), and orthostatic blood pressure]. The results were compared with age matched control materials (finger skin blood flow test, n = 80, and deep-breathing and tilt table tests, n = 56).

RESULTS

The VAC index was found to be significantly increased and the E/I ratio significantly decreased in patients compared to controls, indicating both a sympathetic and a parasympathetic dysfunction. Further, the patients, especially the anti-SSA and anti-SSB antibody seropositives, were found to have an abnormal blood pressure reaction to tilt compared to controls. No correlations were found between autonomic nerve function variables measured and the clinical ophthalmologic or the oral tests, performed at the time of diagnosis.

CONCLUSION

Patients with primary SS show signs of both sympathetic and parasympathetic dysfunction. Further, immunological mechanisms seem to influence blood pressure in patients with primary SS.

Prevalence of fibromyalgia and chronic widespread pain

OBJECTIVE

To explore the prevalence of fibromyalgia and chronic widespread musculoskeletal pain in a general population using the criteria of the American College of Rheumatology from 1990.

DESIGN

Structured interview and clinical examination, including tender-point count and pain threshold measured with a dolorimeter, of subjects with suspected chronic widespread musculoskeletal pain.

SETTING

The general population in south-west Sweden 1995-1996.

SUBJECTS

303 individuals with suspected chronic widespread pain were identified in a previously defined cohort containing 2425 men and women aged 20-74 years. 202 individuals were invited and 147 agreed to participate.

MAIN OUTCOME MEASURES

Tenderpoint count, pain threshold and prevalence of chronic widespread pain and fibromyalgia.

RESULTS

The prevalence of fibromyalgia was estimated to 1.3% (95% CI 0.8-1.7; n = 2425) and that of all chronic widespread pain to 4.2% (95% CI 3.4-5.0; n = 2425). The mean pain threshold measured with a dolorimeter was lower in subjects with chronic widespread pain (p < 0.01) and correlated with the number of tender points (r = -0.59, p < 0.01) but could not be used to distinguish the subjects with fibromyalgia.

CONCLUSION

Compared to other studies, fibromyalgia and chronic widespread musculoskeletal pain seemed to be relatively rare conditions in the south-west of Sweden.

The prevalence of rheumatoid arthritis in Sweden

The aim of this study was to ascertain the prevalence of rheumatoid arthritis (RA) in a Swedish general adult population. A questionnaire about chronic pain was mailed to a total of 3928 subjects who were chosen as a random sample of the population in two communities in the county of Halland. All persons answering affirmatively to questions intended to identify patients with RA were invited to a clinical examination. X-rays of hands and feet, and analyses of rheumatoid factor and C reactive protein were performed provided that the patients fulfilled two or more of the five clinical items of the 1987 ARA criteria. Furthermore, non-participants were searched for in a patient register and in medical records from the local rheumatology unit in an attempt to identify further cases. Using the modified 1987 ARA criteria for population studies the prevalence rate of RA was calculated to 0.51% (95%, CI = 0.31-0.79).

Genealogy construction in a historically isolated population: application to genetic studies of rheumatoid arthritis in the Pima Indian

PURPOSE

Due to the characteristics of complex traits, many traits may not be amenable to traditional epidemiologic methods. We illustrate an approach that defines an isolated population as the "unit" for carrying out studies of complex disease. We provide an example using the Pima Indians, a relatively isolated population, in which the incidence and prevalence of Type 2 diabetes, gallbladder disease, and rheumatoid arthritis (RA) are significantly increased compared with the general U.S. population. A previous study of RA in the Pima utilizing traditional methods failed to detect a genetic effect on the occurrence of the disease.

METHODS

Our approach involved constructing a genealogy for this population and using a genealogic index to investigate familial aggregation. We developed an algorithm to identify biological relationships among 88 RA cases versus 4,000 subsamples of age-matched individuals from the same population. Kinship coefficients were calculated for all possible pairs of RA cases, and similarly for the subsamples.

RESULTS

The sum of the kinship coefficient among all combination of RA pairs, 5.92, was significantly higher than the average of the 4,000 subsamples, 1.99 (p < 0.001), and was elevated over that of the subsamples to the level of second cousin, supporting a genetic effect in the familial aggregation. The mean inbreeding coefficient for the Pima was 0.00009, similar to that reported for other populations; none of the RA cases were inbred.

CONCLUSIONS

The Pima genealogy can be anticipated to provide valuable information for the genetic study of diseases other than RA. Defining an isolated population as the "unit" in which to assess familial aggregation may be advantageous, especially if there are a limited number of cases in the study population.

Survivorship in a population based cohort of patients with Sjögren's syndrome, 1976-1992

OBJECTIVE

Sjögren's syndrome (SS) has been associated with development of lymphoid malignancies and other significant medical complications, but the effect of SS on survival in a population based sample has not been reported. We analyzed survival in an incidence cohort of patients diagnosed with SS in residents of Olmsted County, Minnesota, USA, between 1976 and 1992.

METHODS

All records of physician diagnosed SS were reviewed, as well as all records from patients diagnosed with xerostomia and keratoconjunctivitis sicca, and records of patients with rheumatoid arthritis (RA) and systemic lupus erythematosus. The case definition for SS required 2 of 3 criteria: keratoconjunctivitis sicca, xerostomia, or serologic abnormality. Confounding illnesses were excluded. All patients were white.

RESULTS

Of the 74 cases of SS identified, 50 (67%) had primary SS and 24 (33%) secondary SS. An average of 7.2 years of followup was available for patients with primary SS and 9.9 years for patients with secondary SS. Compared with the general population, patients with SS had increased mortality (p = 0.04). When patients with primary and secondary SS were studied separately, increased mortality was found in patients with secondary SS (p < 0.005) but not primary SS (p = 0.86).

CONCLUSION

In this population based cohort, patients with primary SS did not have increased mortality. However, mortality may have been increased in patients with secondary SS, the majority of whom had RA.

Rheumatoid arthritis in the Pima Indians: the intersection of epidemiologic, demographic, and genealogic data

OBJECTIVE

To describe the clinical features and familial distribution of rheumatoid arthritis (RA) in the Pima Indians.

METHODS

From 1965 through 1990, all cases of RA as defined by the American College of Rheumatology (formerly, the American Rheumatism Association) 1987 criteria or all cases of seropositive, erosive disease as defined by the Rome criteria were identified in individuals who were age 20 years and older and were of 50% or more Pima/Tohono-O'odham heritage. Radiographs were reviewed by 2 musculoskeletal radiologists who were blinded to case status. Kinship coefficients were used to evaluate familial aggregation.

RESULTS

Eighty-eight RA cases were identified from this population-based sample. Over 66% of the cases had seropositive disease, over 60% had erosive disease, and over 40% had subcutaneous nodules. Of the 88 RA cases, 40 were members of families with more than 1 RA case. The remainder were simplex cases.

CONCLUSION

In this population, clinical markers of severe RA were present in a majority of cases. The presence of familial aggregation for RA in the Pima Indians suggests underlying genetic factors in disease pathogenesis.

Evaluation of a structured multidisciplinary day care program in rheumatoid arthritis. A similar effect in newly diagnosed and long-standing disease

The present study evaluated a multidisciplinary structured day-care programme in patients with rheumatoid arthritis of less than 2 years (n = 41) and more than 2 years disease duration (n = 46). During the 3 week intervention, outcome measures reflecting disability (HAQ, SOFI), the patient's perception of disease and pain (VAS for patient's global assessment and pain), Ritchie articular index (RAI), a 44 swollen joint count, and overall disease activity (DAS) improved significantly in the group as a whole. The improvements remained significant after 15 weeks and were of a similar magnitude in the patient groups with short and long disease duration. At week 3 and 15, the ACR and the EULAR criteria for individual response, for the total study group was fulfilled by 28% and 26%, and 36% and 52% respectively. Evaluation of a subgroup 6 weeks prior to admission indicated that the outcome measures were stable at the time of the intervention. Furthermore, administration of intraarticular glucocorticosteroids (GC) could only partly explain the observed improvement. This uncontrolled observational study supports that a multidisciplinary day-care rehabilitation program is beneficial and feasible for patients with rheumatoid arthritis of both short and long duration.

Epidemiology of Sjögren's syndrome, especially its primary form

During the last decades different sets of criteria have been used for defining primary Sjögren's syndrome. Using these criteria prevalence estimates have varied between 0.04%-4.8%. This hundred-fold distribution in estimated frequency is partly due to differences in age groups studied, classification criteria used and methods used for objective evaluation of lachrymal and salivary gland hypofunction. Another problem which makes comparison between studies difficult is that workshop recommendations for the various objective tests are very often not strictly followed. Population based studies tend to identify cases with less severe disease and higher prevalence rates. Furthermore, there seems to be little correlation between dryness symptoms reported by subjects and objective hypofunction of lachrymal and/or salivary glands. Possible explanations to these findings include selection bias towards more severe cases in clinical (university) studies and a low predictive value for some criteria in population studies.

Weight fluctuation, mortality and vascular disease in Pima Indians

OBJECTIVE

To examine the relationship of weight fluctuation to mortality rates and incidence of vascular disease.

SUBJECTS

A cohort of Pima Indians, 572 of whom had non-insulin-dependent diabetes mellitus and 766 without diabetes.

DESIGN

Individuals were invited biennially to research examinations. The root mean square error (RMSE) of the linear trend of weight with time for the first four examinations after age 20 years was used as an index of weight fluctuation. Subjects were followed from the fourth examination until death or until 31 December 1991. The mortality rate ratio (MRR) and its 95% confidence interval (CI) for those with a high weight fluctuation index relative to those with a lower value were determined. The median duration of follow up was 9.3 (range 0.1-22.6) years.

MEASUREMENTS

All cause mortality (n = 356); incidence of diabetic retinopathy (n = 145), diabetic nephropathy (n = 132) and electrocardiographic abnormalities (n = 82).

RESULTS

There was no significant relationship between weight fluctuation and mortality for diabetic subjects (MRR = 1.0, 95% CI 0.8-1.3, p = 0.91). Nondiabetic subjects with a high weight fluctuation index had a higher mortality rate than those with a lower index (MRR = 1.5, 95% CI 1.0-2.1, p = 0.03); the association was stronger among men than among women. The excess mortality in the high weight fluctuation group was not due to cardiovascular diseases, but to noncardiovascular causes and the risk for alcohol-related death was particularly increased. Weight fluctuation was not associated with the incidence of diabetic retinopathy, nephropathy or electrocardiographic abnormalities.

CONCLUSIONS

A high weight fluctuation index was associated with higher mortality rates in nondiabetic, but not in diabetic, Pima Indians. The excess mortality is largely due to noncardiovascular causes of death and may reflect lifestyle factors associated with weight fluctuation, rather than its metabolic effects.

Low prevalences of chronic widespread pain and shoulder disorders among the Pima Indians

OBJECTIVE

To establish the prevalence of shoulder disease and chronic widespread pain in Pima Indians.

METHODS

Cross sectional analyses of data from 4230 subjects for shoulder disease and 105 subjects for chronic widespread pain participating in population surveys

RESULTS

The prevalence of shoulder disease was 4.4% (95% CI, 3.8-5.1), age-sex adjusted to the 1980 US census population. This is lower than in a study of Caucasians [prevalence ratio (PR) = 0.29, 95% CI, 0.20-0.42 for men and PR = 0.55, 95% CI, 0.41-0.73 for women]. Shoulder disease was associated with non-insulin-dependent diabetes mellitus (PR = 1.67, 95% CI, 1.19-2.36). No chronic widespread pain was identified (95% CI, 0-3.5%).

CONCLUSION

Prevalence of these pain syndromes in Pima Indians is lower than in predominantly Caucasian populations. These findings suggest that these populations have different pain perception or different patterns of risk factors for these disorders.

Sjögren's syndrome

The review describes the historical development and nomenclature of Sjögren's syndrome--primary and secondary. The strengths and weaknesses as well as similarities and dissimilarities for establishing the diagnosis of keratoconjunctivitis sicca and xerostomia used among the five most commonly classification criteria sets are focused upon in text and table. Validation of the different sets of criteria are performed. Primary and secondary disease activity markers and outcome measures as well as future research plans for improving diagnosis are dealt with.

The U-shaped association between body mass index and mortality: relationship with weight gain in a Native American population

In order to determine whether weight loss explains high mortality rates in those with a low body mass index (BMI), the relationships between BMI, rate of weight gain and mortality were examined in Pima Indians. Subjects were 814 diabetic and 1814 nondiabetic participants in a longitudinal survey who had at least two examinations after age 20. Median duration of follow-up was 8.1 (range 0.03-25.1) years. BMI showed a U-shaped relationship with mortality rates in men with the lowest rates in the 30-35 kg/m2 category; an inverse relationship was seen in women. Subjects who were losing weight had higher mortality rates than those who were gaining. However, excess mortality among the lightest subjects was present among those who were gaining weight. Among nondiabetic subjects, the mortality ratio (MR) for BMI < 25 kg/m2 compared with 30-35 kg/m2 was 1.5 [95% confidence interval (CI) 1.0-2.2] unadjusted for weight gain, while the adjusted MR was 1.3 [95% CI 0.9-1.9]. Weight loss, which may reflect underlying illness, is associated with high mortality rates in Pima Indians but does not fully account for the high mortality in the lightest individuals.

Comparison of body size measurements as predictors of NIDDM in Pima Indians

OBJECTIVE--To determine and compare the abilities of various anthropometric measurements to predict the development of non-insulin-dependent diabetes mellitus (NIDDM) in Pima Indian men and women. RESEARCH DESIGN AND METHODS--A total of 290 male and 443 female Pima Indians were followed for up to 6 years for the development of NIDDM. A proportional hazards analysis was used to assess the ability of anthropometric measurements evaluated at baseline to predict NIDDM. Receiver operating characteristic (ROC) curves were used to compare individual variables in predicting NIDDM. RESULTS--In separate models controlled for age and sex, body mass index (BMI), waist circumference, thigh circumference, waist-to-thigh ratio (WTR), weight, and percentage body fat (PBF) estimated by bioelectric resistance each predicted NIDDM, which developed in 30 men and 52 women. The highest incidence rate ratios (IRRs; for 1 SD of a variable) were for WTR in men and for PBF in women, although the confidence interval (CI) for PBF was wide. In stepwise analyses, WTR was the most significant predictor in men (IRR for 1 SD = 1.58, 95% CI = 1.20-2.07), and BMI was the most significant predictor in women (IRR for 1 SD = 1.65, 95% CI = 1.29-2.11). However, by ROC analyses, thigh circumference was the only variable significantly worse than WTR in men or BMI in women at predicting NIDDM. CONCLUSIONS-- Measurements such as waist circumference, WTR, weight, and BMI may be useful as more complicated measurements, such as PBF by bioelectrical resistance, for identifying groups of individuals whose body habitus places them at high risk of developing NIDDM.

Rate of weight gain, weight fluctuation, and incidence of NIDDM

The relationships of rate of weight gain and weight fluctuation to incidence of non-insulin-dependent diabetes mellitus (NIDDM) were examined in Pima Indians. The 1,458 subjects were participants in a prospective study with examinations approximately every 2 years. Rate of weight gain was defined as the slope of the regression line of weight with time for two or more consecutive examinations > or = 2 years apart and weight fluctuation as the root-mean-square departure from this line for four examinations. Among men, incidence of NIDDM was strongly and significantly related to rate of weight gain (e.g., age-adjusted incidence = 56.7/1,000 person-years in those with weight gain > or = 3 kg/year and 16.9/1,000 person-years for those losing weight [Ptrend < 0.01]). In women, weight gain was significantly related to diabetes incidence only in those who were not initially overweight (body mass index < 27.3 kg/m2). In contrast to the relationship with weight gain, weight fluctuation was not associated with incidence of diabetes in either sex. These findings suggest that weight control in overweight individuals may be a more effective strategy for prevention of NIDDM in men than in women, whereas prevention of obesity may prevent diabetes in both sexes. Concern about a diabetogenic effect of weight fluctuation should not deter weight-control efforts.

Diabetic nephropathy: a risk factor for diabetes mellitus in offspring

Both non-insulin-dependent diabetes mellitus and diabetic nephropathy show familial aggregation. If diabetes and renal disease have independent determinants (genetic or otherwise), offspring of parents with diabetic renal disease should have a similar risk of diabetes to those offspring of parents with diabetes alone. To test this hypothesis, the prevalence of diabetes was examined in a population-based pedigree study in Pima Indian offspring of three mutually exclusive parental types: 1) diabetic with renal disease, 2) diabetic, but without renal disease and 3) non-diabetic. Among offspring of one diabetic parent and one non-diabetic parent (n = 320) the prevalence of diabetes at ages 15-24 years and 25-34 years was 0% and 11%, respectively if the diabetic parent did not have renal disease compared with 6% and 28% respectively if the diabetic parent did have renal disease. Corresponding rates for offspring of two diabetic parents (n = 121) were 10% and 17%, respectively if neither parent had renal disease compared with 30% and 50%, respectively if one parent did have renal disease. The presence of renal disease in a parent with diabetes relative to diabetes alone was associated with 2.5 times the odds of diabetes (95% confidence interval 1.4-4.3) in the offspring controlled for age, age at onset of parental diabetes and diabetes in the other parent using logistic regression. These findings provide support for parental diabetic renal disease, independent of age at onset of parental diabetes, conferring an increased risk for diabetes in the offspring.(ABSTRACT TRUNCATED AT 250 WORDS)

Meta-analysis reveals association between most common class II haplotype in full-heritage Native Americans and rheumatoid arthritis

The association of RA with the alleles at the HLA system was tested among Pima and Tohono O'odham Indians (Pimans) of the Gila River Indian Community of Arizona. Serologic class I (HLA-A, -B, and -C) alleles were typed in 51 individuals with RA and in 302 without RA. Serologic class II (HLA-DR, DQ; DR52 DR53) alleles were typed in a subset of 47 with RA and 147 without RA. Molecular subtypes of DR3X6, DRB1*1402, and *1406 were determined in 29 individuals, 16 with RA and 13 without RA. Among the cases with RA, 46 of 47 had the serologic antigen HLA-DR3X6, as did 140 of 147 of those without the disease. However, this association was not statistically significant because of the high prevalence of the antigen in the controls. Data from Pimans were analyzed with similar results from the Tlingit and Yakima Indians. A meta-analysis employing the Mantel-Haenszel procedure, stratified by tribe, revealed a statistically significant association between the most common haplotype, DRB1*1402 DQA1*0501 DQB1*0301 DRB3*0101, and RA (summary odds ratio = 2.63, 95% confidence interval = 1.08, 6.46). There was also a statistically significant difference in the genotype distributions of one class I locus, HLA-C, between those with and without RA (chi 2 = 12.4, 5 df; p = 0.03). It is concluded that the association with the most common class II haplotype in full-heritage Native Americans might help explain their high prevalence of RA.

A cross-sectional and longitudinal comparison of the Rome criteria for active rheumatoid arthritis (equivalent to the American College of Rheumatology 1958 criteria) and the American College of Rheumatology 1987 criteria for rheumatoid arthritis

OBJECTIVE

To compare the diagnostic properties of the Rome 1961 criteria for active rheumatoid arthritis (RA) and the American College of Rheumatology (ACR; formerly, the American Rheumatism Association) 1987 criteria for RA with regard to their ability to classify, diagnose, and predict outcome in RA.

METHODS

Analysis of cross-sectional and longitudinal data from repeated health examinations and review of clinical records of 3,509 Pima Indians followed up from January 1966 to December 1990.

RESULTS

The ACR 1987 criteria identified approximately 50% of the cases identified by the Rome 1961 criteria, in both cross-sectional and longitudinal analyses. The ACR 1987 criteria were better predictors of subsequent development of a clinically supported diagnosis and treatment with slow-acting antirheumatic drugs (both P < 0.001), but were less sensitive than the Rome 1961 criteria for detecting cases for which there already was a clinically supported diagnosis (P < 0.001).

CONCLUSION

In a population-based analysis, the ACR 1987 criteria are less sensitive for detecting clinical disease, but predict a clinically more severe prognosis, compared with the Rome 1961 criteria. The sensitivity of both sets of criteria to identify clinical disease is improved if multiple examinations or inactive disease are taken into account.

Decreasing incidence and prevalence of rheumatoid arthritis in Pima Indians over a twenty-five-year period

OBJECTIVE

To evaluate temporal trends in the incidence of rheumatoid arthritis (RA).

METHODS

Incident cases of RA were identified among a population-based cohort of Pima Indians in Arizona over the period 1965-1990.

RESULTS

Among 2,894 subjects, 78 incident cases of RA were identified. The age-adjusted incidence declined by 55% in men (Ptrend = 0.225), and by 57% in women (Ptrend = 0.017) after controlling for oral contraceptive or estrogen use and for pregnancy experience. During the same period, age-adjusted prevalence rates of active RA decreased by 29% in men (Ptrend = 0.63) and by 40% in women (Ptrend = 0.02). Fewer than 17% of subjects with known RA were taking slow-acting antirheumatic drugs (SAARDs) in 1990.

CONCLUSION

The decrease in incidence and prevalence of RA in this population over such a short period implicates the involvement of an environmental factor(s), other than exogenous estrogens, in the pathogenesis of the disease. However, the possibility that the observed decrease might be explained by an increased use of SAARDs in subjects with RA cannot be excluded.

Glucose, insulin concentrations and obesity in childhood and adolescence as predictors of NIDDM

Metabolic abnormalities antedate the development of non-insulin-dependent diabetes mellitus (NIDDM) by some years. How these metabolic abnormalities relate to the genetic component of the disease and to the subsequent prediction of diabetes is unknown. The present study was designed to examine the association of parental diabetes with relative weight, fasting and 2-h plasma glucose and fasting and 2-h serum insulin in childhood, and to identify which of these variables were most predictive of subsequent NIDDM. Subjects comprised 1258 Pima Indians aged 5-19 years with normal glucose tolerance participating in a longitudinal population-based study. Age-sex-adjusted values of relative weight, fasting and 2-h glucose and fasting and 2-h insulin were positively associated with parental diabetes. Only one of 138 subjects with two non-diabetic parents developed diabetes. Among 1120 subjects with at least one diabetic parent, 101 (9.0%) developed diabetes during a mean follow up of 8.4 years. Fasting insulin was a significant predictor of diabetes, but did not add to the predictive value of relative weight. Relative weight and 2-h and fasting plasma glucose were the variables most predictive of NIDDM in childhood and adolescence. Against a background of parental diabetes, high fasting insulin concentrations predict diabetes, compatible with the hypothesis that insulin resistance is an early metabolic abnormality leading to NIDDM. In this study, however, its predictive power did not add significantly to that of relative weight, with which it was correlated.(ABSTRACT TRUNCATED AT 250 WORDS)

Comparison of tests for glycated haemoglobin and fasting and two hour plasma glucose concentrations as diagnostic methods for diabetes

OBJECTIVE

To compare the ability of tests measuring two hour plasma glucose, fasting plasma glucose, and glycated haemoglobin concentrations in predicting the specific microvascular complications of non-insulin dependent diabetes mellitus.

DESIGN

Cross sectional and longitudinal analysis of the relation between complications and concomitant results of the three tests.

SETTING

Gila River Indian Community, Arizona.

SUBJECTS

Pima Indians (cross sectional, n = 960), aged 25 years or above who were not receiving insulin or oral hypoglycaemic treatment at the baseline examination.

MAIN OUTCOME MEASURES

Development of retinopathy and nephropathy.

RESULTS

Cross sectionally, frequency distributions of logarithms of the three sets of results were bimodal, with the prevalence of retinopathy and nephropathy being, respectively, 12.0-26.7 and 3.9-4.2 times as high above as below cut off points which minimised overlap (two hour plasma glucose concentration 12.6 mmol/l; fasting plasma glucose concentration 9.3 mmol/l; glycated haemoglobin (HbA1c) concentration 7.8%). Longitudinally, each of the three measures of glycaemia significantly predicted the development of retinopathy (P < 0.0001) and nephropathy (P < 0.05). Receiver operating characteristic curves showed that two hour plasma glucose concentration was superior to fasting plasma glucose concentration (P < 0.05) for prevalent cases of retinopathy, but otherwise no variable had a significant advantage for detecting incident or prevalent cases of either complication.

CONCLUSIONS

These findings suggest that determination of glycated haemoglobin or fasting plasma glucose concentrations alone may be acceptable alternatives to measuring glucose concentration two hours after challenge with 75 g glucose for the diagnosis of diabetes.

Birth weight and non-insulin dependent diabetes: thrifty genotype, thrifty phenotype, or surviving small baby genotype?

OBJECTIVE

To determine the prevalence of diabetes in relation to birth weight in Pima Indians.

DESIGN

Follow up study of infants born during 1940-72 who had undergone a glucose tolerance test at ages 20-39 years.

SETTING

Gila River Indian community, Arizona.

SUBJECTS

1179 American Indians.

MAIN OUTCOME MEASURE

Prevalence of non-insulin dependent diabetes mellitus (plasma glucose concentration > or = 11.1 mmol/l two hours after ingestion of carbohydrate).

RESULTS

The prevalence was greatest in those with the lowest and highest birth weights. The age adjusted prevalences for birth weights < 2500 g, 2500-4499 g, and > or = 4500 g were 30%, 17%, and 32%, respectively. When age, sex, body mass index, maternal diabetes during pregnancy, and birth year were controlled for, subjects with birth weights < 2500 g had a higher rate than those with weights 2500-4499 g (odds ratio 3.81; 95% confidence interval 1.70 to 8.52). The risk for subsequent diabetes among higher birthweight infants (> or = 4500 g) was associated with maternal diabetes during pregnancy. Most diabetes, however, occurred in subjects with intermediate birth weights (2500-4500 g).

CONCLUSIONS

The relation of the prevalence of diabetes to birth weight in the Pima Indians is U shaped and is related to parental diabetes. Low birth weight is associated with non-insulin dependent diabetes. Given the high mortality of low birthweight infants selective survival in infancy of those genetically predisposed to insulin resistance and diabetes provides an explanation for the observed relation between low birth weight and diabetes and the high prevalence of diabetes in many populations.

Rheumatoid arthritis and mortality. A longitudinal study in Pima Indians

OBJECTIVE

To determine the effect of rheumatoid arthritis (RA) on mortality rates.

METHODS

Longitudinal analyses of data from a cohort of Pima Indians from the Gila River Indian Community in Arizona, who were followed up during the period February 1965 through December 1989.

RESULTS

Among 2,979 study subjects aged > or = 25 years, there were 858 deaths, 79 of which occurred in subjects with RA (36 men, 43 women). Age- and sex-adjusted mortality rates were slightly higher in subjects with RA than in those without (mortality rate ratio 1.28, 95% confidence interval [95% CI] 1.01-1.62). Among those with RA, mortality rates were higher in older subjects (mortality rate ratio 1.51 per 10-year increase in age, 95% CI 1.22-1.88), in male subjects (mortality rate ratio 2.23, 95% CI 1.44-3.45, adjusted for age), and in subjects with proteinuria (mortality rate ratio 1.88, 95% CI 1.02-3.46, adjusted for age and sex). Mortality rate ratios for these risk factors were similar in subjects without RA. In addition, among subjects with RA, rheumatoid factor (RF) positivity was predictive of death (mortality rate ratio 1.94, 95% CI 1.10-3.43), and the excess mortality was found primarily among subjects who were seropositive. The death rate from cardiovascular disease (mortality rate ratio 1.77, 95% CI 1.10-2.84) and from liver cirrhosis or other alcohol-related disease (mortality rate ratio 2.52, 95% CI 1.06-6.01) was increased in persons with RA.

CONCLUSION

The results of this population-based study suggest that although the risk of mortality in subjects with RA is significantly higher than in those without RA, the risk ratio is in the lower range of that described previously in studies of clinic-based cohorts. RF positivity as a predictor of early death among subjects with RA indicates that the immunologic processes in seropositive RA may contribute to the events that eventually lead to early death.

Dry eyes or mouth--an epidemiological study in Swedish adults, with special reference to primary Sjögren's syndrome

The prevalence of dry eyes or dry mouth, and of primary Sjögren's syndrome (primary SS) according to the Copenhagen criteria were established in 705 randomly selected subjects, aged 52-72 years who answered a simple questionnaire, and of whom 247 (35%) reported symptoms. A subgroup with symptoms (n = 77) and a matched asymptomatic control group (n = 32) were examined with the Schimer-1 test (S1t), tear film break-up time (BUT), van Bijsterveld score (vB), unstimulated whole sialometry (Sialo) and, in about 40% of them labial salivary gland (LSG) biopsy and salivary gland scintigraphy. Apart from four cases of keratoconjunctivitis sicca (KCS) among controls, cases of KCS (15), xerostomia (12), autoimmune sialoadenitis (6) and primary SS (6) were exclusively confined to the symtomatic group. The calculated frequencies (with 95% confidence intervals) for the whole population were 14.9 (7.3-22.6)% for KCS, 5.5(3.0-7.9)% for xerostomia and 2.7 (1.0-4.5)% for autoimmune sialoadenitis and primary SS. The serum levels of ANA and RF were similar in the two groups, though those of anti-SS-B/La antibodies were higher in the group with symptoms (P less than 0.01).