Impact of self‐monitoring of salt intake by salt meter in hypertensive patients: A randomized controlled trial (SMAL‐SALT)

Abstract Salt intake over reference level would result in elevated blood pressure (BP) and long‐term morbidity. Salt meter is a device used to detect sodium content in daily food. This study aimed to evaluate the efficacy of salt‐meter addition to dietary education. The authors conducted a randomized‐controlled trial in hypertensive patients with uncontrolled BP (systolic BP ≥140 mmHg or diastolic BP ≥90 mmHg). Patients were randomized to receive salt meter plus dietary education (group A) or education only (group B), and followed up for 8 weeks. The primary endpoint was change in 24‐h urinary sodium excretion. Changes in BP, salt taste sensitivity, cardio‐ankle vascular index (CAVI) were also analyzed. There were total number of 90 patients who had complete follow‐up, 45 in each group. Mean age was 62.9 years and 53% were females. Mean baseline 24‐h urine sodium was 151.6 mmol/24 h and mean SBP and DBP were 152.8 and 83.4 mmHg, respectively. Baseline characteristics were similar between two groups. At 8 weeks, mean change in urine sodium were –31.83 mmol/24 h and 0.36 mmol/24 h in group A and group B, respectively (p = .006). Mean decrease in BP were SBP, 14.44 versus 8.22 mmHg (p = .030), and DBP 5.53 versus 1.93 mmHg (p = .032). The salt sensitivity was improved more in group A. There was no different between change in CAVI. From this study, salt meter in conjunction with dietary education, for self‐monitoring of salt intake is superior to education alone in hypertensive patients, and provided better blood pressure control. Salt meter should be considered in uncontrolled hypertensive patients.


INTRODUCTION
Hypertension is one of the most common chronic medical conditions in clinical practice. According to Health Data Center of Ministry of Public Health, the incidence of newly diagnosed hypertensive patients in Thailand was increasing each year, from 916.89 per 100000 populations in 2014 to 1370.25 per 100000 populations in 2018. 1 However, there was only 30% of patients whose blood pressure was well controlled. 2 Hypertension causes multiple cardiovascular sequelae, for example, coronary artery disease (CAD), heart failure, cerebrovascular accident (CVA), and renal failure. Data from the study of Prospective Studies Collaboration group showed that every 20 mmHg of systolic blood pressure (SBP) or 10 mmHg of diastolic blood pressure (DBP) increment were associated with approximately two times risk of mortality from acute myocardial infarction (MI) and stroke. 3 These consequences impact both public health and economy systems.
High dietary salt intake was believed to be associated with hypertension and was demonstrated in many clinical studies. Each 1-gram (g) sodium intake over the reference level would result in 2.11 mmHg increment in SBP and 0.78 mmHg increment in DBP. 4 In the same way as many studies about salt ingestion and blood pressure in both Western and Asian countries, the decrease in salt intake measured by 24-h urinary sodium excretion was associated with decrease in blood pressure (up to 5.8 mmHg of SBP [5][6][7][8] ) and more pronounced in hypertensive than normotensive populations. 9,10 Furthermore, salt intake over 5 g/day or sodium intake over 2.3 g/day is associated with cardiovascular mortality as mentioned in many studies. 11,12 Dietary salt reduction campaigns have been launched in several countries; however, most of them, especially in low-and middle-income countries, 13 were unsuccessful, in part, due to time limitation in the clinic, lacking of awareness, and the higher threshold to detect salt taste in people with chronic high salt ingestion. From recent national survey, Thai people had consumed more than 9.1 g of salt per day, which was nearly two times above WHO reference level (5 g/day). 14 Sources of excessive salt intake mostly from seasoning, preservative food, and sauce; such as fish sauce, fermented soy sauce, soybean paste, powder bouillon, and pickled fish. Most of all (more than 60%) are in the form of liquid or solution such as curry, soup, or broth. Thus, reduction in consuming seasoning, sauce, or soup would facilitate salt lowering habit. 15 People who regularly consume high sodium diet are less sensitive to salty tasting and have higher salt taste threshold. This factor, in turn, results in the higher sodium intake afterwards. 16

Patient enrollment
Enrolled patients were at least 18 years of age and were diagnosed with hypertension for more than 3 months including either patients with antihypertensive naïve or currently on antihypertensive drug(s

Study objectives
The primary objective was to compare the change in 24-h urine sodium excretion (which represents daily consumption of sodium) from baseline between group of uncontrolled hypertensive patients using salt meter for self-monitoring of salt intake plus dietary education and were also monitored.

Trial conduction
The study was conducted at Ramathibodi Hospital (Bangkok, Thailand) during June 2017 to January 2020. Participants were recruited from outpatient department (family medicine, internal medicine, and cardiology clinics) at Ramathibodi Hospital, Mahidol university (Bangkok, Thailand). All participants were evaluated for eligibility before randomization and follow-up. All patients gave written informed consent before initiation of the protocol.
Hypertensive patients whose blood pressure was uncontrolled were 1:1 randomly assigned, by computer-generated random numbers, to receive salt meter device for self-monitoring of salt intake in addition to dietary education by dietician (group A or device group) or receive only dietary education by dietician (group B or control group).
Each patient was arranged to follow up at 4 and 8 weeks after recruitment and randomization. Patients received brief clinical interview and focused physical examination, measured weight and blood pressure, taken blood for serum sodium, creatinine and eGFR, collected urine for spot urine protein and creatinine at baseline and at 8-week visits. Twenty four hour urine was collected within a day before placing specimen to a laboratory unit to measure 24-h urine sodium excretion at baseline, 4, and 8 weeks ( Figure 1). We estimated glomerular filtration rate (GFR) from serum creatinine based on the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation. 22 Diary form to record use of salt meter was given to each patient in group A to evaluate the adherence and frequency of using device. The blood pressure measurement was done under instruction by the nurse and participants: avoid alcohol, caffeine, and smoking at the day of visit; taking morning antihypertensive drugs 2 h prior to measurement, emptying bladder, sitting for 5-10 min before measurement on a comfortable chair with arm support, and avoid talking while measure the blood pressure. The blood pressure measurement was done three times within 15 min and averaged using a digitalized arm-cuffed blood pressure monitor at the separate unit from usual outpatient clinic to provide noiseless environment. This blood pressure device was calibrated and applied to all of the participants.

24-h urinary sodium collection
Urine samples were collected on the day of completion for determination of urine volume, sodium and creatinine excretion. Urinary sodium was determined using the indirect ion-selective electrode method and the enzymatic method for urine creatinine assay. Urine samples were considered incomplete and discarded if (a) total urine volume was less than 500 ml in 24-h collection; (b) estimated daily urinary creatinine excretion was less than 0.98 g/day for males and less than 0.72 g/day for females 23 ; or (c) reported duration of collection was less than 24 h.

Salt meter devices
We The display screen was developed to display the result in emotion or emotional face graphic that would be easy to understand for all users ( Figure 2A). When the sodium chloride content is in between < 0.7%, 0.7-0.9%, and > 0.9%, there will be smile, poker face, and frown, display respectively ( Figure 2B).

Statistical analysis
We did a primary analysis based on intention-to-treat populations. We calculated the sample size with the n4studies program version 1.4.0. 28 Because there was no current study that demonstrates the efficacy of salt meter device, we assumed the treatment effect of this device for the sample size calculation from the study about effects of health education to the improvement of urine sodium excretion. There was an average of 20% decrement in urinary excretion of sodium 29 ; thence, we used 20% improvement for control group and estimated that there would be 30% improvement in an intervention or salt meter group. The calculated sample size was 45 patients for each group, with 80% power and significant level of 0.05.
For the statistical analyses, we used the SPSS program version 18.0.
Descriptive analysis was done and showed in terms of number, percentage, and mean ± standard deviation (SD). We used Chi-square test for any independent categorical values. For continuous values, we used Student t-test or one-way ANOVA for any independent data. Two-way ANOVA was used to determine interaction between independent factors for additional analysis. All tests were prespecified to the 5% significant level.

RESULTS
A total of 100 patients were enrolled, with 90 patients who met the   Figure 3).
We have analyzed data by two-way ANOVA for two independent variables including intervention-received, stage of hypertension, medication or treatment-naïve, and degree of initial urine sodium excretion, which showed no significant effects or interaction on primary outcome.

DISCUSSION
In uncontrolled hypertensive patients who are on antihypertensive medications or those who are treatment-naïve, high dietary salt intake is considered to be an important contributing factor. Our study   Regulating salt intake behavior in hypertensive patients would finally result in better blood pressure control. Subsequently, better blood pressure control would affect any future morbidity and mortality associated with hypertension. There are many evidences revealed causal relationship of quantity of salt intake and blood pressure. 35,36 Furthermore, a smaller number of antihypertensive agents is another possible benefit in term of health economics and treatment-related adverse reactions.
Because of many possible confounding factors contributing to the blood pressure, we decided to use the less-confounded objective outcome which is urinary sodium excretion as a primary endpoint instead of the blood pressure. With maximal effort to minimize the confounding factors, randomization showed similarity in baseline characteristics data between two groups. In term of blood pressure control, patients in group A or device group have pronounce magnitude in both systolic and diastolic blood pressure lowering; 14.44 and 5.53 mmHg, respectively. These numbers are significantly lower in those of control group with differences of 6.2 mmHg for systolic blood pressure and 3.6 mmHg for diastolic blood pressure. Despite the minimal change in 24-h urine sodium excretion in group B, there was small decrement in blood pressure. We believed that the effect of BP lowering seen in this study was partly from Hawthorne effect. However, there would be Hawthorne effect in both groups, but more pronounced BP lowering in group A could be attributed by salt meter intervention. The effect size in our study is rather larger when compared to DASH diet which reduced systolic blood pressure by 2.8 mmHg and diastolic blood pressure by 1.1 mmHg more than control diet. 37 From meta-analysis of overall effects of salt reduction interventions, including health education and salt substitution, mean effect was -4.3 mmHg in lowering systolic blood pressure and -1.6 mmHg in lowering diastolic blood pressure. 35 Moreover, when compared to antihypertensive monotherapy, results from our study show approximately half an effect to the average of angiotensin-converting enzyme inhibitors which were about 11.5 and 8.6 mmHg lowering in systolic and diastolic blood pressure, respectively. 38 This significant effect may be from the high salt diet consumption and salt-sensitive in elderly patients in the study.
In hypertensive patients, the salt taste sensitivity is worse compared to healthy population 19 and it relates with the preference and familiarity to salty taste. 39 These are obstacles to reduce salt consumption for chronic hypertensive patients and chronic high salt-intake people. With the use of salt meter, patients can differentiate the foods with high content of salt and avoid them. Besides, gradually decrement in amount of salt exceeds the ability of taste bud to distinguish the change. Previous evidence supported that reduction in salt intake for at least 1 week can change the threshold for salty taste. 40 This would be the reason why patients in group A, which had more reduction in salt intake indicated by lower urinary sodium excretion, had better detection threshold than those in group B.
Conversely, this trial could not demonstrate the significant change in CAVI parameter between two groups at the end of follow-up. This might be from the relatively short period of follow-up to observe the vascular change. Longer duration of follow-up may require to see the change in atherosclerotic outcome.
Somehow, there were some limitations of our study such as nonblinded study, since the salt meter device in the intervention arm could not be blinded, neither using placebo in the control arm which might be harmful and unethical. However, we used an objective outcome measurement rather than subjective outcome as well as many approaches to reduce the bias. Dietary education was arranged in the same directions for all participants with same dataset for teaching and certificated dietician who did not know for the patients' allocation. Blood pressure measurement and salt taste sensitivity testing were done with protocol-directed guidelines by the staffs who did not know for the patients' allocation.
Salt meter as a device for self-monitoring of salt intake will facilitate hypertensive patients who have chronic high sodium consumption behavior to be more awareness, self-learning to food with low sodium content, and will gradually improve their salt taste sensitivity. Due to technical design of salt meter that it cannot apply to the solid object or food, some food may be not suitable for testing with salt meter device. Nevertheless, we postulated that use of salt meter would habituate patients to learn which food has high or low content of sodium and subsequently results in eating habit adaptation. Eventually, people may not necessarily need the device for regular monitoring. Lowering salt intake could reduce salt-water retention and regulate the reninangiotensin-aldosterone system which finally result in blood pressure decrement. 36 In conclusions, salt meter in conjunction with dietary education, for self-monitoring of salt intake is superior to education alone in hypertensive patients, and provided better blood pressure control. Improvement in salt taste sensitivity will facilitate the long-term benefit. Salt meter should be considered in patients with uncontrolled hypertension.