Patient characteristics
A total of 353 patients were included in the analysis (142 males and 211 females). Among them, a total of 163 patients (46.2%) had postoperative PHT, and the mean age at surgery was about 44 years. PHT occurred more frequently in men (52.1% vs 30.0%, P<0.001), and in patients with advanced ages (43.0 [43.0-55.5] vs 45.0 [35.0-54.8], P=0.001). In addition, compared with those patients who had normal BP (complete clinical success) after surgery, patients with PHT generally had higher systolic blood pressure (184 [175-200] vs 180 [166-190], P<0.001), hyper BMI values (BMI≥25, 49.1% vs 26.8%, P<0.001), longer duration of hypertension (8.0 [3.0-10.0] vs 3.0 [0.81-7.0], P<0.001), more frequent presence of diabetes mellitus (27.0% vs 5.8%, P <0.001), HHD (34.4% vs 16.8% , P<0.001) and RH (39.9% vs 23.2%, P=0.001), higher DDD of antihypertensive medications (3.6 [2.6-4.4] vs 3.0 [2.5-4.0], P=0.003), lower preoperative ARR (270 [110-541] vs 404 [154-674], P=0.002), lower PAC (293 [192-459] vs 351 [255-533], P=0.006), higher DRC(1.09 [0.50-2.34] vs 0.79 [0.50-1.79], P=0.045), and lower eGFR (96.0 [75.6-109.2] vs 107.2 [91.3-119.5], P<0.001), higher serum potassium nadir (2.70 [2.40-3.00] vs 2.60 [2.16-2.90], P=0.016), and smaller diameter of nodules (1.4 [1.0-1.7] vs 1.5 [1.0-2.0], P=0.017). (table1)
Table1 Preoperative clinical and biochemical characteristics of all patients(training + validation group)
Characteristics
|
Total
|
Persistent hypertension(n=163)
|
Normal blood
pressure(n=190)
|
P
|
Age,y
|
44.0(38.0-55.0)
|
43.0(43.0-55.5)
|
45.0(35.0-54.8)
|
0.001
|
Sex,M/F
|
142/211(38.1%)
|
85/78(52.1%)
|
57/133(30.0%)
|
<0.001
|
SBP,mmHg
|
180(170-200)
|
184(175-200)
|
180(166-190)
|
<0.001
|
DBP,mmHg
|
110(100-120)
|
110(100-120)
|
110(100-120)
|
0.011
|
BMI≥25,yes/no
|
131/222(37.1%)
|
80/83(49.1%)
|
51/139(26.8%)
|
<0.001
|
Duration of HT,y
|
5.0(1.0-10.0)
|
8.0(3.0-10.0)
|
3.0(0.81-7.0)
|
<0.001
|
Family history of HT,yes/no
|
121/232(34.3%)
|
63/100(38.7%)
|
58/132(30.5%)
|
0.136
|
Diabetes,yes/no
|
55/298(21.8%)
|
44/119(27.0%)
|
11/179(5.8%)
|
<0.001
|
HHD,yes/no
|
88/265(24.9%)
|
56/107(34.4%)
|
32/158(16.8%)
|
<0.001
|
RH,yes/no
|
109/244(30.9%)
|
65/98(39.9%)
|
44/146(23.2%)
|
0.001
|
DDD of antihypertensive medication
|
3.6(2.6-4.4)
|
3.6(2.6-4.4)
|
3.0(2.5-4.0)
|
0.003
|
ARR,pg·mL-1/uIU·mL-1
|
340(135-618)
|
270(110-541)
|
404(154-674)
|
0.002
|
PAC,pg/mL
|
326(214-503)
|
293(192-459)
|
351(255-533)
|
0.006
|
DRC,uIU/mL
|
0.96(0.50-2.10)
|
1.09(0.50-2.34)
|
0.79(0.50-1.79)
|
0.045
|
Lowest serum K+,mmol/L
|
2.68(2.20-300)
|
2.70(2.40-3.00)
|
2.60(2.16-2.90)
|
0.016
|
eGFR,(mL/min/1.73m2)
|
101.2(82.9-114.4)
|
96.0(75.6-109.2)
|
107.2(91.3-119.5)
|
<0.001
|
TC,mmol/L
|
4.03±0.83
|
4.08±0.90
|
3.99±0.76
|
0.297
|
TG,mmol/L
|
1.15(0.80-1.63)
|
1.33(0.98-1.89)
|
0.98(0.73-1.40)
|
<0.001
|
Diameter of noduler,cm
|
1.5(1.0-2.0)
|
1.4(1.0-1.7)
|
1.5(1.0-2.0)
|
0.017
|
Surgical side, L/R
|
208/145(58.9%)
|
98/65(60.1%)
|
110/80(57.9%)
|
0.752
|
Total adrenalectomy, yes/no
|
204/149(52.0%)
|
97/66(59.5%)
|
107/83(56.3%)
|
0.619
|
SBP, systolic blood pressure; DBP, diastolic blood pressure; HT, hypertension; HHD, hypertensive heart disease; RH, resistant hypertension; DDD, defined daily dose; ARR, aldosterone/renin ratio; PAC, plasma aldosterone concentration; DRC, direct renin concentration; Lowest serum K+, lowest serum potassium in history; eGFR, estimated glomerular filtration; TC, total cholesterol; TG, triglyceride
Validation group characteristics
There were no significant differences between the training group and the validation group in terms of age, gender, DBP, BMI, duration of hypertension, family history of hypertension, presence of diabetes mellitus, presence of HHD, presence of RH, DDD of antihypertensive medication, ARR, PAC, DRC, lowest serum potassium, eGFR, TC, TG, tumor diameter, and surgical side. (Supplementary table1)
Prediction model
Univariate and multivariate logistic regression analysis were used to determine the predictors of postoperative PHT. Univariate analysis showed that age, gender, SBP, DBP, BMI≥25, duration of HT, presence of diabetes mellitus, HHD, DDD of antihypertensive medication, ARR, PAC, lowest serum potassium, eGFR, and diameter of nodules were all related to PHT after surgery (table2). On the basis of univariate logistic regression analysis, we employed collinearity analysis to eliminate some variables (Supplementary figure). In the multivariate analysis (table3), five variables (BMI≥25, male, diabetes, ARR, duration of hypertension) were all independent risk factors of the PHT.
Table 2 Univariate logistic regression analysis of predictors of hypertension persistence after operation in training group
Variable
|
OR(95% CI)
|
P
|
Age
|
1.03(1.0-1.06)
|
0.013
|
Male
|
2.60(1.53-4.40)
|
<0.001
|
SBP
|
1.02(1.01-1.04)
|
<0.001
|
DBP
|
1.02(1.01-1.04)
|
0.005
|
BMI≥25
|
3.16(1.84-5.42)
|
<0.001
|
Duration of HT
|
1.13(1.07-1.19)
|
<0.001
|
Diabetes
|
6.89(2.74-17.34)
|
<0.001
|
HHD
|
2.42(1.33-4.43)
|
0.004
|
RH
|
2.13(1.24-3.66)
|
0.060
|
DDD of antihypertensive medication
|
1.33(1.12-1.73)
|
0.002
|
ARR
|
0.42(0.23-0.75)
|
0.004
|
PAC
|
0.13(0.04-0.40)
|
<0.001
|
DRC
|
1.06(0.94-1.20)
|
0.347
|
Lowest serum K+
|
1.62(1.05-2.50)
|
0.030
|
eGFR
|
0.98(0.97-0.99)
|
<0.001
|
TG
|
1.14(0.94-1.39)
|
0.178
|
Diameter of nodule
|
0.66(0.46-0.94)
|
0.022
|
Data are expressed as odds ratio (OR)(95% CI). SBP, systolic blood pressure; DBP, diastolic blood pressure; HT, hypertension; HHD, hypertensive heart disease; RH, resistant hypertension; DDD, defined daily dose; ARR, aldosterone/renin ratio; PAC, plasma aldosterone concentration; DRC, direct renin concentration; Lowest serum K+, lowest serum potassium in history; eGFR, estimated glomerular filtration; TG, triglyceride
Table3 Multivariate logistics regression analysis of predictors of hypertension persistence after operation in training group
Variable
|
OR(95% CI)
|
P
|
BMI≥25
|
2.07(1.12-3.81)
|
0.020
|
Male
|
2.33(1.27-4.25)
|
0.006
|
Diabetes
|
3.97(1.44-10.92)
|
0.008
|
ARR
|
0.46(0.23-0.92)
|
0.027
|
Duration of hypertension
|
1.12(1.06-1.19)
|
<0.001
|
Data are expressed as odds ratio (OR)(95% CI). ARR, aldosterone/renin ratio
Finally, these five important variables were included to establish a nomogram (Figure 2). The nomogram showed the score of each variable on each scale. The probability of postoperative PHT was determined by the total score of all variables. According to the model, the ROC curve of the training group and the validation group were drawn. The area under the curve (AUC) of the training group was 0.78 (95%CI 0.72-0.84), and the validation group was 0.77 (95%CI 0.68-0.86) (Figure 3). The C index of the training group and the validation group were 0.783 and 0.769, respectively.
The calibration plot revealed excellent agreement between the probability of PHT estimated by the nomogram and the actual status of PHT (Figure 3). These results showed that the nomogram is efficient in predicting PHT after surgery. DCA was performed in the validation group, and the results showed that when the threshold probability was about 10% to 80%, the nomogram produced a greater net benefit than a treat-all or treat-none strategy, indicating that the nomogram had good clinical value (Figure 4).