All-cause mortality following a cancer diagnosis amongst multiple sclerosis patients: a Swedish population-based cohort study.

BACKGROUND AND PURPOSE: A reduced cancer risk amongst patients with multiple sclerosis (MS) has been reported. Theoretically, this could represent a genuine reduction in risk or, alternatively, 'diagnostic neglect', where cancer is undiagnosed when symptoms are misattributed to MS. OBJECTIVE: Assess all-cause mortality risk following a cancer diagnosis in patients with MS compared with a cohort without MS. PATIENTS: A cohort of MS patients (n = 19,364) and a cohort of the general population (n = 192,519) were extracted from national Swedish registers from 1969 to 2005. All-cause mortality after cancer in MS was compared with the general population. Poisson regression analysis was conducted in the MS and non-MS cohorts separately. The models were adjusted for follow-up duration, year at entry, sex, region and socioeconomic index. The two cohorts were combined and differences in mortality risk were assessed using interaction testing. RESULTS: The adjusted relative risk (and 95% confidence interval) for all-cause mortality following a cancer diagnosis in MS patients (compared with MS patients without cancer) is 3.06 (2.86-3.27; n = 1768) and amongst those without MS 5.73 (5.62-5.85; n = 24,965). This lower magnitude mortality risk in the MS patients was confirmed by multiplicative interaction testing (P < 0.001). CONCLUSIONS: A consistent pattern of lower magnitude of all-cause mortality risk following cancer in MS patients for a range of organ-specific cancer types was found. It suggests that cancer diagnoses tend not to be delayed in MS and diagnostic neglect is unlikely to account for the reduced cancer risk associated with MS. The lower magnitude cancer risk in MS may be due to disease-associated characteristics or exposures.

Multiple sclerosis clinical course and cardiovascular disease risk - Swedish cohort study.

BACKGROUND AND PURPOSE: Cardiovascular disease (CVD) risk amongst multiple sclerosis (MS) patients appears raised, but few studies have examined CVD risk amongst an unselected MS patient group. MS course may be relevant for CVD risk. Our aim was to assess CVD risk and variation by course in MS patients. METHODS: The Multiple Sclerosis Register identified 7667 patients who received an MS diagnosis between 1964 and 2005. They were matched by age, period, region and sex with 76 045 members of the general population without MS using Swedish registers. Poisson regression compared the two cohorts to estimate the relative risk for CVD, overall, as well as grouped and individual CVD diagnoses. RESULTS: MS patients had an increased adjusted relative risk (with 95% confidence intervals; number of MS cohort events) for CVD of 1.31 (1.22-1.41; n = 847), with some variation by course: relapsing-remitting 1.38 (1.17-1.62; n = 168); secondary progressive 1.30 (1.18-1.53; n = 405) and primary progressive 1.15 (0.93-1.41; n = 108). The association for the relapsing-remitting course was not significant after excluding the first year of follow-up. Overall incidence rates per 1000 person-years for CVD are 11.8 (11.06-12.66) for the MS cohort and 8.8 (8.60-9.05) for the non-MS cohort. The most pronounced association was for deep vein thrombosis: relapsing-remitting 2.16 (1.21-3.87; n = 14), secondary progressive 3.41 (2.45-4.75; n = 52) and primary progressive 3.57 (1.95-6.56; n = 15). MS was associated with ischaemic stroke but largely during the first year of follow-up. MS was associated with a decreased relative risk for angina pectoris and atrial fibrillation. CONCLUSIONS: There is a significantly increased relative risk for CVD in MS, particularly for venous thromboembolic disorders in progressive MS, suggesting immobility as a possible factor. An increased frequency of ischaemic stroke in MS is most probably due to surveillance bias resulting from diagnostic investigations for MS. There is no increased relative risk for ischaemic heart disease in MS and atrial fibrillation appears to be less common than amongst the general population.

Shared genetic factors may not explain the raised risk of comorbid inflammatory diseases in multiple sclerosis.

BACKGROUND: Comorbid inflammatory conditions in multiple sclerosis (MS) patients suggest shared risks with MS. OBJECTIVE: To estimate if the risk of immune-mediated disease in MS patients and their parents is increased. METHODS: Swedish register data were analysed using Cox regression to estimate immune-mediated disease risk among 11284 fathers and 12006 mothers of MS patients, compared with 123158 fathers and 129409 mothers of index subjects without MS. Similar analyses were conducted among 20276 index subjects with MS and 203951 without. RESULTS: Parents of patients with MS did not have a significantly altered immune-mediated disease risk. Patients with MS had a consistently raised risk for several immune-mediated diseases: ulcerative colitis, Crohn's disease, type 1 diabetes, psoriasis, polyarthritis nodosa and pemphigoid. The risk was more pronounced for diseases diagnosed subsequent to MS onset. CONCLUSION: The increased occurrence of other immune-mediated diseases in MS patients may not be due to shared genetic factors and surveillance bias is likely to be the main or possibly the entire explanation. If not entirely explained by surveillance bias, a modestly raised occurrence of comorbid diseases may be due to shared environmental risks or factors related to MS disease characteristics.

Appendicectomy and multiple sclerosis risk.

BACKGROUND: Appendicectomy for acute appendicitis, but not for other causes, is inversely associated with immune-mediated diseases such as ulcerative colitis, suggesting appendicitis is a marker of immune characteristics influencing immune-mediated disease risk. This study investigated the association of appendectomy and its underlying diagnosis with multiple sclerosis (MS). METHODS: Swedish general population registers and the Swedish MS register provided information on 20,542 cases with MS diagnosed between 1964-2006 and 204,157 controls matched for age, sex, period and region. Appendicectomy prior to MS diagnosis was identified in 673 cases and 6518 controls. Conditional logistic regression, with adjustment for socio-economic index, assessed the association of diagnosis underlying appendicitis with MS risk. RESULTS: A perforated appendix, the best indicator of acute appendicitis in this material, was inversely associated with MS, although not statistically significantly, with an odds ratio (and 95% confidence interval of 0.86 (0.70-1.04). The odds ratios are 1.04 (0.94-1.16) for appendicitis without perforation and 1.14 (0.98-1.33) for appendectomy without appendicitis. CONCLUSION: Although inconclusive in terms of assessing the hypothesis, these results may help to explain why earlier studies of appendicitis and MS risk have been inconsistent, as there may be variation in association by diagnosis underlying appendicectomy.

Molecular methods for typing of Helicobacter pylori and their applications.

Microbial typing is a useful tool in clinical epidemiology for defining the source and route of infection, for studying the persistence and reinfection rates, clonal selection in the host and bacterial evolution. Phenotypic methods such as biotyping, serotyping and hemagglutinin typing have little discriminatory power compared to genotypic methods concerning the typing of Helicobacter pylori. Therefore great efforts have been made to establish useful molecular typing methods. In this context, the most frequently used genotypic methods are described based on our own experience and the literature: (1) restriction endonuclease analysis, (2) endonuclease analysis using pulsed-field gel electrophoresis, (3) ribotyping, (4) polymerase chain reaction (using either random primers or repetitive DNA sequence primers), and (5) polymerase chain reaction-restriction fragment length polymorphism analysis of e.g. the urease genes. Furthermore, reproducibility, discriminatory power, ease of performance and interpretation, cost and toxic procedures of each method are assessed. To date no direct comparison of all the molecular typing methods described has been performed in the same study with the same H. pylori strains. However, PCR analysis of the urease gene directly on suspensions of H. pylori or gastric biopsy material seems to be useful for routine use and applicable in specific epidemiological situations.