1089 e-Letters

  • RE: High prevalence of hyperuricaemia and its impact on non-valvular atrial fibrillation: the cross-sectional Guangzhou (China) Heart Study

    Lin et al. conducted across-sectional study to evaluate the relationship between the prevalence of hyperuricaemia (HUA) and non-valvular atrial fibrillation (NVAF) (1). Age, living in urban areas, alcohol consumption, central obesity, elevated fasting plasma glucose level, elevated blood pressure, lower high-density lipoprotein cholesterol level and elevated triglycerides level were significantly associated with increased risk of HUA, and HUA were at higher risk for NVAF. Regarding sex difference, serum uric acid level had a modest predictive value for NVAF in women. I have some concerns about their study.

    First, Tu et al. conducted a prospective study to evaluate the risk of gout in patients with alcohol-related diseases and alcohol dependence syndrome (2). Alcohol-related diseases were significantly associated with gout risk, and severe alcohol-dependent patients were significantly associated with an increased risk of gout. The authors recommended that alcohol use assessment and measures to prevent alcohol dependence in patients with gout. I agree with their recommendation, which would partly lead to prevent NVAF.

    Second, Yu et al. presented strategies to improve gout/hyperuricemia management for preventing acute arthritis attacks, cardiovascular disease, kidney disease, and other urate-related disorders (3). Alcohol consumption is closely related to gout/hyperuricemia, and psycho-social factors should also be considered for the management.

    Third, Ku...

    Show More
  • Competing risks should be accounted for in the assessment of the risk of fetal death

    It has been five years since Hedegaard and colleagues published their paper on stillbirth and labour induction [1]. In Denmark, its findings are still often brought forward as weighty documentation in public and professional discussions on the appropriate timing of recommended induction of labour in low-risk pregnancies after due date. Thus, as late as in January 2020, one of the main Danish broadcasting companies (TV2) covered the subject, and a curve almost identical to the top graph of the study’s Figure 2 was presented, alongside statements from one the authors. The graph was entitled "Number of dead fetuses per 1000 pregnant women for the period 2000-2008", and the accompanying text read “[…] the numbers show that the number of dead fetuses increases markedly by the beginning of 41 completed gestational weeks and even more after 42 weeks. […] the number of fetuses that died in the mother’s womb after 37th week decreased from 120 to 60 when the regulations were changed in 2011.”

    Following its publication in 2014, the study was subject to much attention from peers. The responses focused on data quality, methodological issues, and whether data did or did not support the authors’ conclusions. However, the problem regarding competing risks was not touched upon, and since ignoring this problem will make the risk appear greater than it actually is, we find it crucial to point out that competing risks was not taken into account in this study’s assessment of...

    Show More
  • RE: Guideline-directed medical therapy in elderly patients with heart failure with reduced ejection fraction: a cohort study

    Seo et al. conducted a prospective study to investigate the risk of 3-year all-cause mortality in elderly patients with heart failure with reduced ejection fraction (HFrEF) with special reference to the clinical characteristics and treatment strategies (1). Adjusted hazard ratio (HR) (95% CI) of guideline-directed medical therapy (GDMT) for all-cause mortality was 0.47 (0.39 to 0.57). In addition, adjusted HR (95% CI) of beta-blockers only and renin-angiotensin system inhibitors only for all-cause mortality was 0.57 (0.45 to 0.73) and 0.58 (0.48 to 0.71), respectively. I have a query about their study with special emphasis on sex difference.

    Motiejunaite et al. conducted a prospective study to evaluate sex difference in 1-year all-cause mortality in patients with acute heart failure (AHF) (2). In the GREAT registry mostly from Europe and Asia, adjusted hazard ratio (HR) (95% CI) of women against men for 1-year mortality was 0.86 (0.79-0.94), which was obvious in northeast Asia, presenting 0.76 (0.67-0.87). In the OPTIMIZE-HF registry from the USA, adjusted HR (95% CI) of women against men for 1-year mortality was 0.93 (0.89-0.97). They concluded that male patients with AHF had an increased mortality risk than female patients. Although they handled patients with AHF, the follow-up period and GDMT should be paid attention for the mortality risk assessment. In contrast, Seo et al. should conduct stratified analysis by sex.

    Regarding the first query, Lainscak et...

    Show More
  • Fibroids and Obstetrical Outcomes

    To the Editor,

    I read with interest the article by Karlsen, et al. concluding that here was an association between clinically significant uterine fibroids and pre-term birth. (1) The study has several significant shortcomings. The likely inaccuracies of the use of a large retrospective, administrative, coded database and the low prevalence of fibroids only identified clinically and not confirmed by ultrasound or other imaging modality were mentioned by the authors in the limitations section of the article. However, the degree of these limitations makes the data uninterpretable and the conclusions unfounded.

    Administrative databases are inadequate to answer clinical questions as they are subject to clinical misdiagnosis and coding errors which are blind to the goals of the study. The only way to assure accuracy of such a data set is to adjudicate the data by chart review, which was not done by the authors.

    Reliance on clinical examination to determine the presence of uterine fibroids, especially submucous fibroids that might more significantly impact pregnancy, is inaccurate. (3) Virtually all women in the developed world currently have an ultrasound as part of their early pregnancy evaluation. Most studies using ultrasound for the diagnosis of fibroids in early pregnancy report a prevalence of 3-4%. One such study found that among 64,047 pregnant women, 2058 had fibroids diagnosed with 1st trimester US, a prevalence of 3.2%. (Stout)

    The c...

    Show More
  • RE: Association between depressive symptoms and arterial stiffness: a cross-sectional study in the general Chinese population

    Peng et al. conducted a cross-sectional study to determine the relationship between depressive symptoms and arterial stiffness in general population (1). The Patient Health Questionnaire-9 (PHQ-9) was used to assess the degree of depressive symptoms: 0-4 no depressive symptoms, 5-9 mild depressive symptoms and 10-27 moderate to severe depressive symptoms. Brachial-ankle pulse wave velocity (baPWV) was measured to determine arterial stiffness. By multivariate linear regression analysis, standardized regression coefficients (95% CIs) of mild and moderate to severe depressive symptoms to predict baPWV were 40.3 (6.6-74.1) and 87.7 (24.0-151.5), respectively. In addition, there were significant interactions in subjects whose blood pressures are beyond the optimal range and combined with diabetes mellitus. Two major metabolic components, increased blood pressure and glucose intolerance, were closely associated with the relationship between depressive symptoms and arterial stiffness. I have two concerns about their study.

    First, Liu et al. examined the association between arterial stiffness and type 2 diabetes mellitus (T2DM) with special reference to interaction by white blood cell (WBC) counts (2). They observed a dose-response relationship between increased baPWV and elevated risk of T2DM, presenting the adjusted odds ratio (OR) (95% CI) of highest tertile of baPWV for T2DM risk being 2.29 (1.32-3.98). In addition, mediation analyses indicated that total WBC count medi...

    Show More
  • A cluster randomised trial of cloth masks compared with medical masks in healthcare workers

    The suggestion that cloth masks can lead to increased infection compared to no mask is not substantiated The control arm had less than 1% of no mask use and therefore the statement is an assumption not a proof . It could be possible for bacteria or fungi that could multiply on the mask but for virus it would have to shown there is more bioavailability than no mask .

  • Vicki Macleod

    The authors do not state what kind / standard of cloth masks were used. Is there any chance they could do this. Given the huge shortage of FFP2 and FFP3 masks, it would be helpful to know the authors’ views of advice from the CDC in the US. https://www.cdc.gov/coronavirus/2019-ncov/prevent-getting-sick/diy-cloth...

  • RE: Circulating liver enzymes and risks of chronic diseases and mortality in the prospective EPIC-Heidelberg case-cohort study

    Katzke et al. conducted a prospective study to assess the relationships of alkaline phosphatase (ALP), gamma-glutamyltransferase (GGT), alanine aminotransferase (ALT), aspartate transaminase (AST) and the De Ritis ratio (AST/ALT) with mortalities (1). There were significant positive associations of all-cause mortality with ALP, GGT and AST. In addition, a combined liver risk score had positive associations with all-cause and cause-specific mortality. I have two concerns about their study.

    First, Oh et al. conducted a cross-sectional study to investigate the association between renal hyperfiltration and serum ALP level (2). Renal hyperfiltration was defined as exceeding the age- and sex-specific 97.5th percentile, and odds ratio (95% confidence interval [CI]) of the highest ALP quartile for renal hyperfiltration was 1.624 (1.204-2.192). Higher ALP levels are significantly associated with renal hyperfiltration, and renal hyperfiltration was closely associated with several metabolic disorders. As renal hyperfiltration can be considered as early-stage of renal damage, serum ALP level might be related to renal damage. Mechanism of the association should be specified by further study.

    Second, Kunutsor et al. conducted a meta-analysis to evaluate the associations of baseline levels of liver enzymes with all-cause mortality in general populations (3). The pooled relative risks (95% CI) per 5 U/l increment in GGT and ALP levels were 1.07 (1.04-1.10) and 1.03 (1.01-1.0...

    Show More
  • RE: Prevalence and factors associated with fatigue in the Lausanne middle-aged population: a population-based, cross-sectional survey

    Galland-Decker et al. conducted a cross-sectional study to assess factors associated with fatigue in the middle-aged general population (1). Adjusted odds ratios (ORs) (95% confidence intervals [CIs]) of obesity, depression, and anaemia for fatigue were 1.40 (1.03-1.91), 3.26 (2.38-4.46), and 1.70 (1.00-2.89), respectively. I have two concerns about their study with special reference to sex difference and sleep parameters.

    First, Broström et al. conducted a cross-sectional study to examine the associations between self-reported sleep duration, depressive symptoms, anxiety, fatigue and daytime sleepiness in older inhabitants (2). Subjects were divided into short sleepers (≤6 hours), normal sleepers (7-8 hours), and long sleepers (≥9 hours). Depressive symptoms were associated with short sleep in men, and fatigue was associated with both short and long sleep duration in men. As there is a significant association between sleep duration, depressive symptoms and fatigue only in men, mechanism of sex difference on the association should be specified by prospective studies, including sex-related variables.

    Second, Lang et al. examined the effect of co-morbid insomnia on depression in community-dwelling middle-aged men (3). Insomnia was defined as difficulty initiating/maintaining sleep (DIMS). Co-morbid insomnia with daytime fatigue/OSA was defined as DIMS-F/COMISA. Standardized regression coefficients (95% CIs) of DIMS-F and COMISA for depression, using Patient Hea...

    Show More
  • Dear Dr. Karlsen and Colleauges,

    We read Dr. Karlsen’s paper entitled “Relationship between a uterine fibroid diagnosis and the risk of adverse obstetrical outcomes: a cohort study” with great interest. We understand the authors evaluated a large and complicated set of data and appreciate their effort. However, the authors used a database fraught with errors without validation of said database, made incorrect assumptions and conclusions about the data, and incompletely discussed some of the major queries in the discussion.

    The first issue surrounds the use of the Danish National Birth Cohort, Patient Registry and Birth Registry. Although the authors write about the limitations of a registry that uses codes “more closely connected to hospital budgets than hospital diagnosis codes”, they rely on a single study to prove the quality of their database.

    Multiple authors have shown that administrative databases are inadequate to answer clinical questions. They are are set up for billing purposes only, not epidemiological research (1,2), and elaborate statistical machinations do not overcome the impact of anomalous data (3). Several studies have shown that the very registry used in this report; the Danish National Patient Registry, is highly inaccurate from a clinical standpoint. The authors from a study in 2009 have a self-proclaimed 10% inaccuracy rate in their diagnoses addressing oral contraceptive research (4). Other authors have found higher erroneous rates for diagnoses of hypertension, rhe...

    Show More