On the power of Chatterjee’s rank correlation

Chatterjee (2021) introduced a simple new rank correlation coefficient that has attracted much attention recently. The coefficient has the unusual appeal that it not only estimates a population quantity first proposed by Dette et al. (2013) that is zero if and only if the underlying pair of random variables is independent, but also is asymptotically normal under independence. This paper compares Chatterjee’s new correlation coefficient with three established rank correlations that also facilitate consistent tests of independence, namely Hoeffding’s $D$, Blum–Kiefer–Rosenblatt’s $R$, and Bergsma–Dassios–Yanagimoto’s $\tau^*$. We compare the computational efficiency of these rank correlation coefficients in light of recent advances, and investigate their power against local rotation and mixture alternatives. Our main results show that Chatterjee’s coefficient is unfortunately rate-suboptimal compared to $D$, $R$ and $\tau^*$. The situation is more subtle for a related earlier estimator of Dette et al. (2013). These results favour $D$, $R$ and $\tau^*$ over Chatterjee’s new correlation coefficient for the purpose of testing independence.

Assessment of vision-related quality of life and depression and anxiety rates in patients with neovascular age-related macular degeneration

PURPOSE

To assess the vision-related quality of life and the depression and anxiety rates in patients with neovascular Age-Related Macular Degeneration (nAMD).

METHODS

A cross-sectional study of patients with nAMD treated with intravitreal injections was performed. The patients completed two validated questionnaires: the Visual Functioning Questionnaire (VFQ-25, score from 0 to 100), and the Hospital Anxiety and Depression Scale (HADS) questionnaire. Age, gender and visual acuity (VA) in the Early Treatment Diabetic Retinopathy Study (ETDRS) scale was registered.

RESULTS

Fifty-five patients with nAMD participated with a mean age of 80.9 ± 6.6 years-old (range 67-93) and a mean VA in the best eye of 73.5 ± 12.7 letters (range 44-95). The global VFQ-25 mean score was 57.4 ± 21.9 being 38.9 ± 13.2 for the general vision and 42.0 ± 19.5 for the general health. VA in the best eye was associated with the global score of the VFQ-25 scale (R = 0.608; P < .001), but no correlation was observed with general health (P = .936). In the HADS scale, 26.9% and 25.5% of patients had symptoms of depression and anxiety respectively. A negative correlation was found between the HADS and VFQ-25 scales for the general vision score (R = -0.438).

CONCLUSION

This study elucidates the impact of vision impairment and the visual functioning in nAMD, describing an important rate of depression and anxiety symptoms.

POS0817 A NOVEL MODEL TO ASSESS DISEASE ACTIVITY IN TAKAYASU ARTERITIS BASED ON 18F-FDG-PET/CT: A CHINESE COHORT STUDY

Background:

Takayasu arteritis (TA) is a condition characterized by major large-vessel vasculitis (LVV), and is most commonly found in young women (age <40 years) of East Asia countries. 18F-FDG-PET/CT has been widely used in the diagnosis and follow-up of cancers to gather functional information based on metabolic activity. In the present study, we evaluated the value of different parameters in 18F-FDG-PET/CT for assessing active TA disease, and we establish a simple, quantifiable, and effective disease activity evaluation model based on 18F-FDG-PET/CT. A comparison in the ability to identify active disease was performed between the established Kerr score and the new 18F-FDG-PET/CT was also performed.

Objectives:

To investigate the utility of 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG-PET/CT) in assessing disease activity in TA.

Methods:

Ninety-one patients with TA, were recruited from a Chinese cohort from October 2017 to January 2019. Clinical data, acute-phase reactants (APRs), and 18F-FDG-PET/CT findings were simultaneously recorded. The Physician Global Assessment was used as the gold standard to assess TA disease activity. The value of using 18F-FDG-PET/CT to identify active disease was evaluated, using erythrocyte sedimentation rate (ESR) as a reference. Disease activity assessment models were constructed and concordance index (C-index), net reclassification index (NRI), and integrated discrimination index (IDI) were evaluated to compare the benefits of the new modes with ESR and Kerr score.

Results:

In total, 64 (70.3%) cases showed active disease. Higher levels of ESR and CRP, and lower interleukin (IL)-2R levels, were observed in active cases. 18F-FDG-PET/CT parameters, including SUVmean, SUVratio1, SUVratio2, sum of SUVmean, and sum of SUVmax, were significantly higher in active disease groups. The C index threshold of ESR to indicate active disease was 0.78 (95% CI: 0.69-0.88). The new activity assessment model combining ESR, sum of SUVmean, and IL-2R showed significant improvement in C index over the ESR method (0.96 vs. 0.78, P < 0.01; NRI 1.63, P < 0.01; and IDI 0.48, P < 0.01). The new model also demonstrated modest superiority to Kerr score assessment (0.96 vs. 0.87, P = 0.03; NRI 1.19, P < 0.01; and IDI 0.33 P < 0.01).

Conclusion:

A novel 18F-FDG-PET/CT-based method that involves combining the sum of SUVmean with ESR score and IL-2R levels demonstrated superiority in identifying active TA compared to conventional methods.

References:

[1]Kerr GS, Hallahan CW, Giordano J, Leavitt RY, Fauci AS, Rottem M, et al. Takayasu arteritis. Ann Intern Med 1994;120:919-29.

[2]Hoffman GS, Ahmed AE. Surrogate markers of disease activity in patients with Takayasu arteritis. A preliminary report from The International Network for the Study of the Systemic Vasculitides (INSSYS). Int J Cardiol 1998;66 Suppl 1:S191-4; discussion S195.

[3]Misra R, Danda D, Rajappa SM, Ghosh A, Gupta R, Mahendranath KM, et al. Development and initial validation of the Indian Takayasu Clinical Activity Score (ITAS2010). Rheumatology (Oxford) 2013;52:1795-801.

[4]Bardi M, Diamantopoulos AP. EULAR recommendations for the use of imaging in large vessel vasculitis in clinical practice summary. Radiol Med 2019;124:965-972.

[5]Spick C, Herrmann K, Czernin J. 18F-FDG PET/CT and PET/MRI Perform Equally Well in Cancer: Evidence from Studies on More Than 2,300 Patients. J Nucl Med 2016;57:420-30.

Disclosure of Interests:

None declared

POS0650 PREDICTORS OF DURABLE CLINICAL RESPONSE TO TOFACITINIB 11 MG ONCE DAILY WITH OR WITHOUT METHOTREXATE IN PATIENTS WITH RHEUMATOID ARTHRITIS: POST HOC ANALYSIS OF DATA FROM A PHASE 3b/4 METHOTREXATE WITHDRAWAL STUDY

Background:

ORAL Shift, a global Phase 3b/4 non-inferiority study, demonstrated sustained efficacy and safety of tofacitinib modified-release (MR) 11 mg once daily (QD) following methotrexate (MTX) withdrawal in patients with rheumatoid arthritis (RA) who achieved Clinical Disease Activity Index (CDAI) low disease activity (LDA) after treatment with tofacitinib + MTX.1

Objectives:

To assess predictors of durable clinical response in patients receiving tofacitinib MR 11 mg QD in ORAL Shift.

Methods:

ORAL Shift (NCT02831855) enrolled patients aged ≥18 years with moderate to severe RA and an inadequate response to MTX. Patients received open-label tofacitinib MR 11 mg QD + MTX for 24 weeks. Patients achieving LDA (CDAI score ≤10) at Week (W)24 entered the 24-week double-blind MTX withdrawal phase and were randomised 1:1 to receive tofacitinib MR 11 mg QD + placebo (tofacitinib monotherapy; ie blinded MTX withdrawal) or continue tofacitinib + MTX. In this post hoc analysis of randomised patients, we assessed predictors of durable response (maintenance of response from W24–48) per CDAI LDA and remission (CDAI score ≤2.8) criteria. All covariates were initially assessed for significance in a univariate logistic regression. Highly correlated covariates were reviewed to assess which would be removed prior to modelling in a multivariable logistic regression. Remaining significant (p≤0.10) covariates in the univariate regression were selected in the model using a stepwise selection process with p≤0.15 entry and p≤0.05 stay criteria. From the final model, estimated odds ratios (ORs) with 95% confidence intervals (CIs) are presented.

Results:

In the double-blind phase of ORAL Shift, durable CDAI LDA and remission rates were: 66.2% and 14.7%, respectively, with tofacitinib + MTX (N=266); and 55.3% and 11.0%, respectively, with tofacitinib + placebo (N=264) (Table 1). In the multivariable analysis, five patient covariates significantly predicted durable CDAI LDA (Figure 1; discussed hereafter). Each unit increase in CDAI score at W24 reduced the likelihood of maintaining CDAI LDA by 22.0%. Each unit increase in C-reactive protein (CRP) at W24 increased the likelihood of maintaining CDAI LDA by 4.0%; this may have been due to imbalanced CRP levels at W24 (randomisation) between treatment groups (Figure 1, footnote c). The odds of durable CDAI LDA were 53.0% lower in the US vs Europe and 61.0% lower in the US vs ‘other’ regions. Each unit increase in baseline Health Assessment Questionnaire-Disability Index (HAQ-DI) score reduced the odds of durable CDAI LDA by 34.0%. Patients receiving tofacitinib + MTX had 66.0% greater odds of durable CDAI LDA vs patients receiving tofacitinib + placebo. CDAI at W24 was the only significant predictor of durable CDAI remission in the multivariable analysis: OR (95% CI) 0.32 (0.24, 0.43); p<0.0001. Each unit increase in CDAI score at W24 reduced the odds of durable CDAI remission by 68.0%.

Table 1.

Durable CDAI LDA and remissiona in patients receiving tofacitinib MR 11 mg QD with MTX or placebo in the double-blind phase of ORAL Shift

Tofacitinib + MTX(N=266)Tofacitinib + placebo(N=264)Durable CDAI LDA, n (%)176 (66.2)146 (55.3)Durable CDAI remission, n (%)39 (14.7)29 (11.0)

aDurable CDAI LDA or remission was defined as achievement of LDA (CDAI score ≤10) or remission (CDAI score ≤2.8), respectively, at W24–48

N, number of patients in each group; n, number of patients achieving outcome

Conclusion:

This post hoc analysis of data from ORAL Shift found that CDAI and CRP at W24, geographic region, baseline HAQ-DI and treatment could be predictors for durable CDAI LDA. As these findings were limited to patients who achieved CDAI LDA at W24 with tofacitinib MR 11 mg QD + MTX, additional data in the general patient population need to be investigated.

References:

[1]Cohen et al. Lancet Rheumatol 2019; 1: E23-34.

Acknowledgements:

Study sponsored by Pfizer Inc. Medical writing support was provided by Sarah Piggott, CMC Connect, and funded by Pfizer Inc.

Disclosure of Interests:

Kunihiro Yamaoka Speakers bureau: Actelion, Astellas, Chugai, Eisai, Eli Lilly, GlaxoSmithKline, Janssen, Mitsubishi Tanabe, Nippon Shinyaku, Pfizer Inc, Takeda, Consultant of: Actelion, Astellas, Chugai, Eisai, Eli Lilly, GlaxoSmithKline, Janssen, Mitsubishi Tanabe, Nippon Shinyaku, Pfizer Inc, Takeda, Stanley B. Cohen Consultant of: AbbVie, Eli Lilly, Genentech, Gilead Sciences, Pfizer Inc, Grant/research support from: AbbVie, Eli Lilly, Genentech, Gilead Sciences, Pfizer Inc, Naonobu Sugiyama Shareholder of: Pfizer Inc, Employee of: Pfizer Inc, Harry Shi Shareholder of: Pfizer Inc, Employee of: Pfizer Inc, Jose Luis Rivas Shareholder of: Pfizer Inc, Employee of: Pfizer Inc, Annette Diehl Shareholder of: Pfizer Inc, Employee of: Pfizer Inc, Josef S. Smolen Consultant of: AbbVie, AstraZeneca, Celgene, Celltrion, Chugai, Eli Lilly, Gilead Sciences, ILTOO, Janssen, Novartis-Sandoz, Pfizer Inc, Roche, Samsung, Sanofi, Grant/research support from: AbbVie and AstraZeneca

POS0086 ANALYSIS OF DISEASE ACTIVITY MEASURES IN THE CONTEXT OF A METHOTREXATE WITHDRAWAL STUDY AMONG PATIENTS WITH RHEUMATOID ARTHRITIS TREATED WITH TOFACITINIB 11 MG ONCE DAILY + METHOTREXATE: POST HOC ANALYSIS OF DATA FROM ORAL SHIFT

Background:

The Phase 3b/4 study ORAL Shift demonstrated sustained efficacy and safety of tofacitinib modified-release (MR) 11 mg once daily (QD) following methotrexate (MTX) withdrawal that was non-inferior to continued tofacitinib + MTX use (per DAS28-4[ESR]), in patients (pts) with rheumatoid arthritis (RA) who achieved CDAI-defined low disease activity (LDA) with tofacitinib + MTX at Week (W)24.1

Objectives:

To assess the performance of alternative disease activity measures at W24 (randomisation) and W48 (study endpoint) in ORAL Shift.

Methods:

ORAL Shift (NCT02831855) enrolled pts aged ≥18 years with moderate to severe RA and an inadequate response to MTX. Pts received open-label tofacitinib MR 11 mg QD + MTX for 24 weeks. Achievement of CDAI LDA (≤10) at W24 was set as the criteria for entry to the 24-week double-blind MTX withdrawal phase, with pts randomised 1:1 to receive tofacitinib MR 11 mg QD + placebo (PBO) (ie blinded MTX withdrawal) or continue tofacitinib + MTX. In this post hoc analysis, efficacy analyses were performed in 8 subgroups defined by achievement of various disease activity criteria at W24: DAS28-4(ESR) remission (<2.6) or LDA (≤3.2); DAS28-4(CRP) <2.6 or ≤3.2; RAPID3 remission (≤3) or LDA (≤6); CDAI remission (≤2.8); and SDAI remission (≤3.3). For each subgroup, the proportion of pts who achieved the corresponding disease activity criterion at W48 was calculated, with a 95% confidence interval (CI) estimated using the normal approximation to the binomial distribution. The change (Δ) from W24 to W48 in least squares (LS) mean DAS28-4(ESR) and DAS28-4(CRP) was also calculated in each subgroup, with a 95% CI for the difference between treatment groups estimated using a mixed model with repeated measures. Nominal p values were calculated and are presented with no formal statistical hypothesis testing formulated.

Results:

Overall, 694 pts entered the open-label phase of ORAL Shift, and 530 were randomised and received treatment in the double-blind phase; 264 and 266 pts received tofacitinib + PBO and tofacitinib + MTX, respectively (Figure 1a). Considering those pts who were randomised and treated, the proportion of pts achieving each disease activity criterion at W24 varied, but was similar between treatments within each subgroup (Figure 1a). Among pts who met each disease activity criterion at W24, generally the majority of pts in both treatment groups also met the same criterion at W48 (Figure 1b). Numerically more pts receiving tofacitinib + MTX vs tofacitinib + PBO continued to meet the corresponding criterion at W48. Regardless of the disease activity criterion met at W24, differences between treatment groups in LS mean ΔDAS28-4(ESR) (Figure 1c) and ΔDAS28-4(CRP) (data not shown) from W24 to W48 favoured tofacitinib + MTX vs tofacitinib + PBO.

Conclusion:

This post hoc analysis of data from pts randomised and treated in ORAL Shift demonstrated that, regardless of the disease activity state criterion met at W24, generally a majority of pts receiving tofacitinib maintained achievement of the corresponding disease activity criterion at W48, with or without continued MTX. Differences between treatment groups in LS mean ΔDAS28-4(ESR) from W24 to W48, as defined by achievement of LDA or remission with a variety of disease activity measures, were less than a change of 1.2, which is considered to be the threshold for a minimal clinically important improvement.2

References:

[1]Cohen et al. Lancet Rheumatol 2019; 1: E23-34.

[2]Ward et al. Ann Rheum Dis 2015; 74: 1691-1696.

Acknowledgements:

Study sponsored by Pfizer Inc. Medical writing support was provided by Gemma Turner, CMC Connect, and funded by Pfizer Inc.

Disclosure of Interests:

Roy Fleischmann Speakers bureau: Pfizer Inc, Consultant of: AbbVie, Amgen, Bristol-Myers Squibb, Celltrion, Eli Lilly, GlaxoSmithKline, Janssen, Novartis, Pfizer Inc, Sanofi-Aventis, UCB, Grant/research support from: AbbVie, Amgen, AstraZeneca, Bristol-Myers Squibb, Celltrion, Eli Lilly, Genentech, GlaxoSmithKline, Janssen, Novartis, Pfizer Inc, Samumed, Sanofi-Aventis, UCB, VORSO, Boulos Haraoui Speakers bureau: Amgen, Pfizer Inc, UCB, Consultant of: AbbVie, Amgen, Eli Lilly, Merck, Pfizer Inc, UCB, Grant/research support from: AbbVie, Maya H Buch Speakers bureau: AbbVie, Consultant of: AbbVie, Eli Lilly, Gilead, MSD, Pfizer Inc, Roche, Sanofi, Grant/research support from: Pfizer Inc, Roche, UCB, David Gold Shareholder of: Pfizer Inc, Employee of: Pfizer Inc, Gosford Sawyerr Consultant of: Pfizer Inc, Employee of: Syneos Health Inc, Harry Shi Shareholder of: Pfizer Inc, Employee of: Pfizer Inc, Annette Diehl Shareholder of: Pfizer Inc, Employee of: Pfizer Inc, Kristen Lee Shareholder of: Pfizer Inc, Employee of: Pfizer Inc.

[Genetics Analysis of patients with Dravet syndrome due to mosaicism variation of paternal SCN1A gene]

Genetic analysis was performed on a family of fraternal twins affected with Dravet syndrome by genetic tests whose parents were normal. To further analyze the cause of the disease, the fraternal twins were subjected to whole exome sequencing (WES), and the family was verified by Sanger sequencing, with the father semen and peripheral blood DNA were further analysed by target sequencing. The WES test identified a heterozygous c.5348C>T (p.Ala1783Val) variant of the SCN1A gene in the fraternal twins, which was predicted to be pathogenic and was detected in the father peripheral blood and semen, but not in the mother. So the mosaicism mutation of paternal SCN1A gene might be the genetic cause of Dravet syndrome in offspring.

Distance to water source in early childhood affects growth: a cohort study

OBJECTIVES

Approximately 165 million children aged <5 years are at risk of low height-for-age. This study explored whether water quality and distance to water source affected short- and long-term growth, and the critical time windows of such effects.

STUDY DESIGN

This study used data from the China Health and Nutrition Survey (CHNS), which was a cohort study performed in 15 Chinese provinces.

METHODS

Data from the CHNS between 1989 and 2011 were examined. Three cohorts of children who were enrolled at 0-2 years of age in 1989, 1993, and 2000 provided sufficient data for this study. Child height was measured by physicians, and household water supply, including the water quality (safe or unsafe) and distance to the source (in-yard or out-yard), was assessed using a questionnaire when the children were 0-2, 4-6, and 11-13 years of age. Multiple regressions were performed to analyse the associations between water quality and distance to the source at various ages, and height at that age and older ages, with and without adjusting for the household water supply at previous ages. Multiple informant models were created using a generalised estimating equation and these were used to assess whether the exposure coefficients were equal across the three age periods.

RESULTS

A total of 1192 children were included. Water quality was not significantly associated with height over the three age periods. An out-yard water source at the age of 0-2 years was not associated with child height at that age, but it was inversely associated with child height at ages 4-6 and 11-13 years (β = -1.605, 95% confidence interval [CI]: -2.490 to -0.720, and β = -2.817, 95% CI: -4.411 to -1.224, respectively), after adjusting for sociodemographic and economic covariates and baseline child height. Distance to the water source at age 4-6 and 11-13 years was not significantly associated with height. However, significant differences were observed in the associations between distance to water source across the three age periods and height at age 4-6 and 11-13 years (p_int = 0.064 and 0.069, respectively).

CONCLUSIONS

Distance to the water source in early, but not later, childhood exerted a long-term effect on child height. When the water quality is good, efforts should be made to shorten the distance to the source to improve child health and growth.

[Consistency of effective orifice area of prosthetic mitral valve estimated using 2-dimensional and 3-dimensional transesophageal echocardiography]

OBJECTIVE

To analyze the consistency of effective orifice area (EOA) of prosthetic mitral valve estimated using 2- dimensional (2D) and 3-dimensional (3D) transesophageal echocardiography (TEE).

OBJECTIVE

This study was conducted among 34 patients undergoing mitral valve replacement surgery in Nanjing First Hospital between March and June in 2019. The diameter of the left ventricular outflow tract (LVOT) measured by 2D-TEE was used to calculate the cross sectional area of LVOT (CSALVOT). In 3D-TEE method, LVOT area was measured directly by planimetry on an enface view. The EOAs of the prosthetic mitral valve were calculated for both methods using the continuity equation. Bland-Altman plot consistency test was used to analyze the consistency between the two sets of EOA results, and linear regression analysis was used to analyze their correlation.

OBJECTIVE

The EOA of the prosthetic mitral valve differed significantly between 2D method and 3D method (2.22±0.71 cm² vs 2.35±0.70 cm², P < 0.001) with a mean difference of -0.14±0.20 cm² and 95% coherence boundaries of (-0.53, 0.25 cm²). The regression equation for EOA-3D and EOA-2D is y=0.27 + 0.94x, showing a good correlation between the two methods.

OBJECTIVE

EOA estimation of the prosthetic mitral valve using 2D and 3D TEE has a good consistency, and the results estimated by the 2D method are slightly lower by about 6% than those by the 3D method.

Adrenal metastases: early biphasic contrast-enhanced CT findings with emphasis on differentiation from lipid-poor adrenal adenomas

AIM

To evaluate the accuracy of unenhanced attenuation and early biphasic contrast-enhanced computed tomography (CT) in differentiating adrenal metastases (AMs) from lipid-poor adrenal adenomas (AAs).

MATERIALS AND METHODS

This retrospective study included 37 patients with 50 AMs and 86 patients with 89 lipid-poor AAs. Quantitative data including the longest diameter (LD), the shortest diameter (SD), LD/SD ratio, CT attenuation values (CTu, CTa, CTv), degree of enhancement (DEAP, DEPP, DEpeak, APW, RPW), and peak enhanced/unenhanced (PE/U) CT attenuation ratio were obtained. Qualitative data including enhancement pattern, location, shape, the presence of calcification or haemorrhage, and intra-lesion necrosis were analysed. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were also calculated.

RESULTS

The PE/U ratio (≤1.25), CTu (≥32.2 HU), DEpeak (≤43.15 HU), DEPP (≤37.65 HU), presence of intralesional necrosis, location (bilateral adrenal glands), and irregular shape were significant variables for differentiating AMs from lipid-poor AAs (p<0.05). Among them, PE/U ratio (≤1.25) was of greater value in differentiating the two adrenal diseases, with sensitivity, specificity, area under the receiver operating curve (ROC) curve (AUC) of 92%, 84%, 0.933, respectively. When at least any three of above criteria were combined, the sensitivity, specificity, PPV, and NPV for diagnosing AMs were 88%, 93%, 88%, and 88%, respectively.

CONCLUSIONS

These seven CT criteria are conducive to differentiate AMs from lipid-poor AAs. Early biphasic contrast-enhanced CT is a high-efficient and practical imaging tool in differentiating them.

Association of C-Reactive Protein and Motoric Cognitive Risk Syndrome in Community-Dwelling Older Adults: The China Health and Retirement Longitudinal Study

OBJECTIVES

Motoric cognitive risk syndrome (MCR) is a newly described pre-dementia syndrome characterized by cognitive complaints and slow gait and is associated with numerous adverse outcomes. Previous studies have indicated an association between C-reactive protein (CRP) and cognitive decline, but no clear relationship between CRP and MCR has been reported. The purpose of the study is to examine the associations between CRP with MCR and MCR subtypes.

METHODS

Participants were 5,642 adults aged ≥60 years from the China Health and Retirement Longitudinal Study (CHARLS). MCR was defined as cognitive complaints and slow gait speed without dementia or impaired mobility. Two subtypes of MCR were defined by whether memory impairment (MI) was also present, such as MCR-MI and MCR-non-MI. MI was evaluated through the immediate recall and delayed recall in a word recall test during the CHARLS and was defined as 1.0 standard deviation or more below the mean values of the test scores in this cohort.

RESULTS

Of the participants, 421 (7.46%) met the criteria for MCR. After multivariate adjustment, participants with higher CRP levels had an increased likelihood of MCR (fourth quartile: adjusted odds ratio [OR]=1.44; 95% confidence interval [CI]: 1.06-1.95) compared with those in the first quartile group. The OR for MCR-MI was 2.04 (95% CI: 1.35-3.09) for the highest quartile of CRP compared to the lowest quartile. No significant associations between CRP levels and odds of MCR-non-MI were observed.

CONCLUSIONS

Higher CRP levels were associated with increased odds of prevalent MCR-MI but not MCR-non-MI among community-dwelling older adults.

Association between cigarette smoking and the susceptibility of acute myeloid leukemia: a systematic review and meta-analysis

OBJECTIVE

To elucidate the association between cigarette smoking and the susceptibility of acute myeloid leukemia (AML).

MATERIALS AND METHODS

We searched relevant articles from PubMed, Embase, and Cochrane by 1st December, 2018. This meta-analysis included 20 case-control studies, involving 7,538 AML patients and 137,924 healthy controls. Studies reported OR and 95%CI of the correlation between cigarette smoking and AML susceptibility were eligible. Subsequently, the included data were weighted by an inverse variance and analyzed using fixed-effects or random-effects model. Subgroup analysis was conducted based on ethnicities and sources of controls. Heterogeneity test was applied for the included articles. Data analyses were conducted using STATA 12.0.

RESULTS

Current smokers (OR=1.42, 95%CI= 1.28-1.57; p=0.392) and ever-smokers (OR=1.16, 95%CI=1.05-1.28; p=0.036) were associated with AML susceptibility. In the subgroup analysis by ethnicity, only current smokers (OR=1.45, 95%CI=1.29-1.63; p=0.371) and ever-smokers (OR=1.16, 95%CI=1.03-1.30; p=0.034) of the Caucasian population were associated with AML susceptibility. Stratified analysis based on SOC (source of controls) indicated increased susceptibility of AML in current smokers (OR=1.43, 95%CI=1.26-1.63; p=0.283) and ever-smokers (OR=1.20, 95%CI=1.07-1.35; p=0.078) of the population-based group. Nevertheless, only current smokers in the hospital-based group had increased susceptibility of AML (OR=1.45, 95%CI=1.19-1.76; p=0.198). The risks of AML in ever-smokers (OR=1.04, 95%CI = 0.81-1.35; p=0.054) of the hospital-based group did not remarkably changed.

CONCLUSIONS

In this meta-analysis, we confirmed the association between cigarette smoking and onset risk of AML, especially in the Caucasian population. High-quality, large-scale researches are required to be conducted in multi-center hospitals for verification.