Demographic characteristics of patients and controls
The effects of confounding factors such as sex and age were excluded from the present study. Table 1 summarized the mean age of patients was 64.6 ± 6.7 years, whereas that of the controls was 64.0 ± 7.0 years. The two groups exhibited no significant difference in terms of age and sex (p > 0.05).
OP prevalence in patients and controls
Table 1 showed a significant difference was observed in the bone mass between the EH group and the control group (χ2 = 12.968, p = 0.002). (figure 1:Comparison of bone mass between EH group and control group)The percentage of people with normal bone mass in the EH group was 28.0% (119/425), whereas that in the control group was 39.3% (167/425). The incidence of OP was 30.4% (129/425) in the EH group and 23.3% (99/425) in the control group. This difference was statistically significant (χ2 = 12.968, p = 0.002) even after controlling for sociodemographic confounding factors by using logistic regression. Additionally, the incidence of low BMD (osteopenia + osteoporosis) was 72.0% (306/425) in the EH group and 60.7% (258/425) in the control group, and this difference also remained statistically significant after adjusting for sociodemographic confounding variables (p < 0.01). BMD levels observed in the EH group were significantly lower than those in the control group (U = 83167.000, p = 0.046). Additionally, no significant sex predilection was observed in the prevalence of OP in the two groups (χ2 = 1.998, df = 1, p = 0.157), however, female patients in the EH group exhibited a significantly higher prevalence of low BMD (osteoporosis + osteopenia) (male patients: 57.1%, female patients: 71.6%, χ2 = 7.855, df = 1, p = 0.005), although this difference was not observed in the control group (χ2 = 0.455, df = 1, p = 0.500). Additionally, a significant sex predilection was observed in the BMD T-score among the patients (male patients: −1.19 ± 1.55, female patients: −1.70 ± 1.34, F = 6.142, df = 423, p = 0.001), which persisted even after controlling for confounding factors (p < 0.01). Further analysis exhibited that although the body weight of patients with EH was significantly higher than that of healthy controls (EH patients: 66.15 ± 10.11, healthy controls: 62.23 ± 9.21, t = 5.760, p < 0.001), the CR levels between the two groups did not vary significantly (EH group: 70.68 ± 0.88, healthy controls: 69.69 ± 0.72, p = 0.934).
Bone densitometry examination of BMD in the lumbar spine and hip indicated that the prevalence of OP and low BMD in the EH group was significantly higher than that in the control group. Additionally, BMD levels were significantly lower in patients with EH than in healthy controls. No sex predilection was observed in the prevalence of OP in both the EH and control groups. However, the prevalence rate of low BMD in the EH group was significantly higher in female patients than in male patients, whereas no such difference was observed in the control group. Additionally, female patients exhibited significantly lower BMD T-scores than male patients.
Demographic and clinical variables in the non-OP and OP populations of the EH group
According to Table 2, a comparison of the OP and non-OP populations in the EH group indicated the prevalence of OP in 32 of 112 men (28.6%) and 97 of 313 women (31.0%) . Additionally, the analysis of cross-tabulation results exhibited that the OP risk was similar between middle-aged and elderly male and female patients, with no statistically significant difference (Pearson χ2 = 0.228, p = 0.633). However, a linear correlation was observed between age and OP incidence (χ2 = 8.991, p = 0.001). Pearson correlation results (R = 0.146 and p = 0.003) indicated that the OP incidence increased with age. The body weight of the non-OP population (65.86 ± 9.84 Kg) was similar to that of the OP population (66.80 ± 10.71 Kg), with no statistically significant difference (t = 0.874, p = 0.383). A total of 92 of the 329 participants (28.0%) with no fracture history and 37 of the 96 (38.5%) patients with previous fracture exhibited OP. This result in combination with the cross-tabulation results indicated that the OP risk was higher in the population with previous fractures (χ2 = 3.934, p = 0.047). With regards to the results of the investigation on OP risk with nighttime urination frequency, OP incidence was noted in 20 of the 102 (19.6%) individuals who never got up at night for urination, 79 of the 238 (33.2%) individuals who urinated 1–2 times at night, and 30 of the 85 (35.3%) individuals who urinated >3 times at night, with statistically significant differences among the three groups (χ2 = 7.461, p = 0.024). A two-by-two comparison revealed a statistically significant difference in the OP risk between those who never got up at night and those who urinated 1–2 times and >3 times at night (p < 0.05), whereas no significant difference was observed in the OP risk between those who urinated 1–2 times at night and those who urinated >3 times at night (p > 0.05). The survey on depression and anxiety in EH patients revealed that OP occurred in 84 of the 315 patients (26.7%) who were not depressed and anxious, 38 of the 95 patients (40.0%) who exhibited mild depression and anxiety, and 9 of the 14 patients (64.3%) who exhibited moderate depression and anxiety. Fisher's exact test exhibited that the OP risk was not equal among the study subjects with different depressive and anxiety states, indicating that the OP risk was statistically different in at least two groups (p = 0.001). A two-by-two comparison revealed a statistically significant difference in the OP risk between those who never got up at night and those who urinated 1–2 times and >3 times at night (χ2 = 6.208, p = 0.013). However, no difference was observed in the OP risk between those who urinated 1–2 times at night and those who urinated >3 times at night (p > 0.05). Based on the duration of persistence of EH, the patients were divided into four groups: 2–5 years, 6–10 years, 11–19 years, and more than 20 years. The incidence of OP was noted in 32 of the 158 patients (20.3%) with 2–5 years of disease, 49 of the 137 patients (35.8%) with 6–10 years of disease, 21 of the 52 patients (40.4%) with 11–19 years of disease, and 27 of the 78 patients (34.6%) with more than 20 years of disease. The difference between the four groups was statistically significant (χ2 = 12.669, p = 0.005). A two-by-two comparison revealed a statistically significant difference in the OP risk between those with 2–5 years of disease and those with 6–10 years of disease (p < 0.05). However, no difference in the OP risk was observed among patients with more than 5 years of disease.
The OP risks were similar in middle-aged and elderly men and women, with no significant differences. However, age and OP exhibited a linear correlation, and the incidence of OP increased with age. Additionally, the body weight of the non-OP population was approximately similar to that of the OP population. However, the OP risk was higher in those with previous fractures. Other risk factors for OP identified were: age, fracture history, nocturnal urination frequency, depression and anxiety status, duration of hypertension, and antihypertensive medication.
OP in patients treated with different antihypertensive medications
In the EH group, the proportion of patients who received only one antihypertensive medication was 89.2% (264/296) in the non-OP population, whereas that in the OP population was 10.8% (32/296). The proportion receiving a combination of two or three antihypertensive medications was 81.4% (105/129) in the non-OP population and 18.6% (24/129) in the OP population. The present study demonstrated a statistically significant difference in the OP risk between patients on a combination of antihypertensive medications and those on a single antihypertensive medication (χ2 = 4.770, df = 1, p = 0.029). Furthermore, the EH patients were divided into seven groups based on the type of antihypertensive medication used.According to Table 3, OP occurred in 17 of 30 (56.7%) individuals who were not on regular medication, 77 of 202 (38.1%) individuals who were on oral amlodipine, and 8 of 20 (40.0%) individuals who were on oral nifedipine, 5 of 32 (15.6%) individuals who took oral propranolol, 5 of 31 (36.1%) individuals who took oral captopril, 9 of 80 (11.3%) individuals who took oral valsartan, and 8 of 30 individuals who took proprietary Chinese medicines (pCMs) (specific ingredients unknown). The difference in OP incidence between the seven groups was statistically significant (χ2 = 36.722, p < 0.000). A two-by-two comparison revealed statistically significant differences in the OP risk in patients who did not take the drug regularly and those who took oral propranolol, captopril, valsartan, and oral pCMs (p < 0.05). However, no significant differences were observed in the OP risk in patients who did not take the drug regularly and those who took oral amlodipine or nifedipine. The difference in the OP risk between patients on oral amlodipine or nifedipine and those on oral propranolol, captopril, and valsartan was statistically significant (p < 0.05).(figure 2:Two-by-two comparison based on the χ2 test for different antihypertensive medication applications).
This study showed that the combined treatment of antihypertensive drugs increased the incidence of OP compared with EH subjects who only used one type of antihypertensive drug alone. In the comparison of different types of antihypertensive drugs taken by EH patients, it is found that the risk of OP in patients taking propranolol, captopril, and valsartan is relatively lower than other types of antihypertensive drugs.
Effect of differences in EH and OP prevalence on CR
The Mann–Whitney U test was used to determine whether a difference in CR levels existed between the non-OP and OP populations in the EH group (Table 2). The histogram exhibited an inconsistent shape of the distribution of CR levels in the two groups. The mean rank order of CR levels in the non-OP group was 225.17, whereas that in the OP group was 185.07. The Mann–Whitney U test exhibited a statistically significant difference in CR levels between the two groups (U = 15489.000, p = 0.002). Further, the two-factor ANOVA exhibited that the data in this study had no outliers, the residuals were close to normal distribution (p > 0.05), and all variables exhibited equal variance (p > 0.05). In terms of CR value, an interaction was observed between EH prevalence conditions and bone mass conditions (F = 3.584, p = 0.028, bias η2 = 0.008). The separate effects analysis suggested that the effect on CR values differed among the study subjects with different bone mass conditions and different EH prevalence conditions: the EH study subjects (F = 6.346, p = 0.002, bias η2 = 0.015), and between study subjects with different EH prevalence conditions in different bone mass conditions: the population with normal bone mass (F = 10.997, p = 0.001, bias η2 = 0.013).
No significant difference was observed in the CR levels between the EH and control groups, and CR concentrations were significantly higher in the non-OP population than in the OP population within the EH group. An interaction between different EH prevalence and different bone mass conditions had an influence on CR values. Moreover, serum CR levels in the study subjects with different bone mass conditions were more affected by the presence or absence of EH. The expression of CR levels varies significantly under the interaction of EH, and the application of CR in patients with EH may seriously reduce the diagnostic efficacy of OP.