Patients enrollment and study design
The cross-sectional study was conducted on 505 gout patients who visited the Shandong Province Gout Clinic, the Affiliated Hospital of Qingdao University from August 2015 to October 2019 and performed joint ultrasound examination, which mainly analyzed via electronic medical records, and called back the unclear parts of the medical records. All patients satisfied the 2015 ACR/ EULAR classification criteria for gout . Exclusive criteria were: (1) Patients with a history of arthritis other than gout; (2) Patients with a history of trauma or surgery at the ultrasound site; (3) Patients who experienced acute gout attacks during the examination; (4) Patients unable to cooperate with examination and medical history collection. This study finally included 356 patients.
We retrieve the electronic medical records of all subjects and record their age, gender, disease duration, serum alanine aminotransferase (ALT), serum aspartate aminotransferase (AST), transaminase ratio(AST/ALT), blood urea nitrogen (BUN), serum creatinine (Cr), the estimated glomerular filtration rate (eGFR) , Body Mass Index (BMI), systolic blood pressure (SBP), diastolic blood pressure (DBP), serum uric acid (UA), fasting blood glucose(GLU), serum triglyceride (TG), serum cholesterol (CH), urate-lowering therapy (ULT), family history of gout, history of alcohol and ultrasound examination results of all patients . The estimated glomerular ltration rate (eGFR) was calculated from the CKD-EPI equation: GFR = 141 × min(Scr/κ, 1)α × max(Scr/κ, 1)-1.209 × 0.993Age × 1.018 [if female] _ 1.159 [if black], where Scr is serum creatinine, κ is 0.7 for females and 0.9 for males, α is -0.329 for females and -0.411 for males, min indicates the minimum of Scr/κ or 1, and max indicates the maximum of Scr/κ or 1. All subjects received standard clinical and laboratory examinations and evaluations. The study was approved by the ethics committee of Affiliated Hospital of Qingdao University, and obtained the informed consents of all participants.
Assessment of alcohol history
At the Clinical Medical Center of Gout of the Affiliated Hospital of Qingdao University, the trained doctors conducted systematic interviews or telephone return calls to all participants to determine their history of alcohol. Data on history of alcohol include: frequency of drinking (times/ week), alcohol intake (g/week), drinking time (years), type of drinking (beer/ liquor/ wine). The type of drinking was recorded as the most frequently consumed alcoholic beverage. Alcohol intake (g/ week): According to the ratio recommended in the "Dietary Guidelines for Chinese Residents" (2007): 25g alcohol (ethanol) is equivalent to 750mL beer, 250mLwine, and 50g liquor, we determined the alcohol intake each time via multiplying the consumption of each alcoholic beverage by its ethanol content, and then multiplied by the frequency of drinking (times/ week) to calculate the average weekly alcohol intake. Participants were divided into three categories according to their alcohol intake status: non-drinkers, former drinkers, and drinkers. People who used to drink alcohol frequently in the past, but have not drunk in the past year before the ultrasound examination were defined as former drinkers .
Assessment of potential confounders
Alcohol consumption and tophi may be affected by various other conditions. Therefore, we systematically assessed the potential confounding factors of all participants. Previous studies have shown that in addition to drinking, age, disease course, blood pressure, eGFR, UA and ULT are also risk factors of tophi[30, 31]. We analyzed whether there were differences in the above indicators between patients in different groups, and conducted independent risk factor analyses of tophi. As shown in Supplementary Table 1, age, duration of gout, SBP, DBP, TG, BUN, Cr, eGFR, family history of gout and alcohol consumption are independent risk factors for tophi. Hence, we included the above indicators except alcohol consumption as confounding factors into the subsequent regression analyses.
The ultrasound examination was performed by two experienced sonographers who were blind to each other's diagnosis and the subjects’ clinical information. Sonographic examinations were performed using the ALOKA 70 ultrasound system (HITACHI) with a 9-13 MHz linear transducer. Ultrasound examination was performed on the affected joint/joints of each subject, which was/were the most frequent or most painful areas of gout attack. A total of 427 joints were examined on 356 subjects, and the examination sites include the first to fifth metatarsophalangeal (MTP) joints, ankle, knee, calcaneus, as well as metacarpal (MCP) joints, wrist and elbow joints. All the above joints were explored on both transverse and longitudinal planes from the dorsal, palmar, medial and lateral aspects. Besides, all joint areas were examined by ultrasound in a standardized manner.
A total of five ultrasound signs were observed and recorded during the inspection (Fig. 1), as shown below:(1) Tophi was an inhomogeneous substance with poorly defined margins, low to high echo, single or clustered, sometimes accompanied by calcification or surrounding low echo halo (Fig. 1a and Fig. 1b) [32, 26]. (2) Effusion was defined as an abnormal echo-free space existing in or around the joint cavity, without color Doppler signals (Fig. 1c). (3) Synovial hypertrophy appeared as abnormally low echo tissue in the joint space, and were concentric thickening of the synovium (Fig. 1d). (4) According to the OMERACT standard, bone erosion was defined as the discontinuity of the hyperechoic bone surface profile in two perpendicular planes (Fig. 1e) . (5)Double contour sign (DCS) was defined as an irregular hyperechoic band on the edge of articular cartilage surface, which could be continuous or intermittent, regardless of the angle of ultrasound (Fig. 1f) [26, 28]. All ultrasound features were defined as the present or absent. During the ultrasound examination, we recorded the presence or absence of the above five signs in the affected joints, the number (recorded as "none", "single ", "multiple") and size [recorded as the maximum diameter (cm)] of tophus.
The characteristics of the subjects were described by simple descriptive statistics: mean ± standard deviation [SD] or median (25th 75th percentiles) for the continuous variable and the frequency of the categorical variable (%). Quantitative data to the normal distribution were compared using the 1-way analysis of variance, while for variables not following the normal distribution and ordered categorical variable, the non-parametric test (Kruskal-Wallis test) were used to compare the differences between the groups. Unordered categorical variables were analyzed using the chi-squared test (χ2test) and Fisher’s exact test. In order to explore the relationship between alcohol consumption and ultrasound signs such as tophi in gout patients, we performed regression analyses as follows. To begin with, we performed multiple regression analyses with the history of alcohol as independent variables and the five ultrasound signs as dependent variables. In our analyses, no drinker was used as the reference in each model to compare the effect of alcohol consumption relative to no drinking. Moreover, we tested two models to control for potential confounders that might affect the association between alcohol consumption and ultrasound signs. The first model (model 1) did not contain any covariates and the second model (model 2) included age, sex, duration of gout, SBP, DBP, BUN, Cr, eGFR, UA, ULT, Family history of gout as covariates. Subsequently, a multiple linear and logistic regression were conducted to explore the specific effects of alcohol consumption on the size and number of tophus, meanwhile controlling for the same covariates.
What’s more, to investigate the effects of age, duration of gout and ULT on the association between alcohol consumption and the size of tophus in above analyses, a similar regression analysis was repeated including a 2-way interaction term between history of alcohol and age, duration of gout and ULT as additional independent variables, the size of tophus as dependent variable. Subsequently, further subgroup analyses were conducted based on the results of above analyses.
Tests were two-tailed and P <0.05 was considered statistically significant. All data were analyzed using SPSS Statistics version 25.0.