General information
A total of 100 patients received therapy from June 2014 to December 2020. There were 53 patients in the PTX group, of which 47 (88.7%) underwent total PTX with AT; 6 (11.3%) underwent total PTX, and a total of 47 patients in the RFA group, of which 26 (55.0%) performed single-session RFA; 21 (45%) received two-session RFA.
The baseline characteristics of the patients are summarized in Table 1. The average age of these patients was 50.61±12.41 years old. The average dialytic vintage was 7.77±3.60 years. The median follow-up time was 30.025 (22.325-38.725) months. And 81% patients were undergoing hemodialysis. There was no statistical difference between the two groups in several variables such as age, gender, dialysis history, follow-up time, renal function, UA, ALB, Hb, CRP, TNI, BNP, iPTH, calcium, phosphorus, bALP, etc. (P>0.05).
More parathyroid hyperplasia nodules were detected in the PTX group. There were 3.92±0.43 nodules resected in the PTX group, 3.68±0.63 nodules ablated in the RFA group (P=0.027). But there was no difference in the nodule’s maximum diameter and preoperative clinical symptoms (ostealgia or arthralgia, cutaneous pruritus, skeleton distortion, calcinosis cutis) between the two groups. The results of bone-derived turnover markers, osteoporosis, and vascular calcification suggested the existence of chronic kidney disease related mineral and bone disorder(CKD-MBD) in both groups.
Table 1. Patients’ baseline characteristics.
Parameter
|
PTX Group(n=53)
|
RFA Group(n=47)
|
P value
|
Age(years)
|
50.36±12.73
|
50.89±12.17
|
0.831
|
Genger, male
|
66.0%
|
61.7%
|
0.652
|
Dialysis method, Hemodialysis
|
83.0%
|
78.7%
|
0.585
|
Dialytic vintage (years)
|
7.85±3.76
|
7.69±3.45
|
0.824
|
Follow-up time (months)
|
31.70(23.40-42.70)
|
28.60(21.30-36.60)
|
0.085
|
Nodule numbers
|
3.92±0.43
|
3.68±0.63
|
0.027
|
Nodule’s maximum diameter (mm)
|
19.18±5.45
|
20.31±5.54
|
0.328
|
Creatinine(μmol/L)
|
883.53±251.86
|
847.30±216.04
|
0.445
|
Uric acid (μmol/L)
|
418.36±118.53
|
416.72±103.97
|
0.942
|
Albumin(g/L)
|
36.74±4.64
|
36.39±4.18
|
0.698
|
Haemoglobin (g/L)
|
107.36±18.68
|
101.02±23.01
|
0.132
|
CRP(mg/L)
|
7.12±10.48
|
11.12±14.90
|
0.195
|
TnI(μg/L)
|
0.04±0.05
|
0.04±0.04
|
0.870
|
BNP(pg/ml)
|
158.85(73.85-422.5)
|
230.80(73.65-646.75)
|
0.316
|
Baseline iPTH (pg/mL)
|
1857.47±812.34
|
1747.31±924.09
|
0.527
|
Calcium (mmol/L)
|
2.47±0.18
|
2.48±0.24
|
0.884
|
Phosphate(mmol/L)
|
2.33±0.53
|
2.22±0.59
|
0.331
|
bALP(U/L)
|
132.31±33.83
|
146.07±62.65
|
0.246
|
β-CTx(pg/ml)
|
5405.12±951.37
|
5449.51±843.51
|
0.818
|
N-MID(ng/ml)
|
289.69±391.48
|
242.34±56.58
|
0.451
|
tPINP(ng/ml)
|
1104.22±296.67
|
1072.29±260.13
|
0.595
|
25(OH)D(ng/ml)
|
21.82±10.49
|
20.96±13.40
|
0.730
|
Preoperative clinical symptoms, yes
|
62.3%
|
53.2%
|
0.359
|
Osteoporosis, yes
|
38.1%
|
42.3%
|
0.706
|
Carotid atherosclerosis, yes
|
92.3%
|
76.2%
|
0.123
|
Abbreviations: PTX, Parathyroidectomy; RFA: Radiofrequency Ablation; CRP, C-reactive protein; TnI, Troponin I; BNP, B-type natriuretic peptide; iPTH, intact parathyroid hormone; bALP, bone-specific alkaline phosphatase; β-CTx, Beta C-terminal cross-linked telopeptides of typeⅠcollagen; N-MID, N-terminal osteocalcin; tPINP, total type I collagen N-terminal propeptide
Outcomes
Primary outcomes: At the time of discharge, 12.0% of patients in the PTX group and 32.6% of the patients in the RFA group had iPTH concentrations within the target range (124-558pg/mL) (P=0.015). The majority of patients in both groups had low iPTH concentrations (<124pg/mL). In addition, the iPTH concentrations of 25.8% of patients in the PTX group and 51.3% of the RFA group achieved the recommended goal at the study endpoint (P=0.031). 58.1% of patients in the PTX group who didn’t achieve target values still had low iPTH concentrations. (Table 2)
Table 2. Proportion of patients achieving and not achieving the target range iPTH concentration by treatment groups.
iPTH level (pg/mL)
|
PTX Group
|
RFA Group
|
P value
|
Discharge after treatment
|
|
|
0.051
|
124-558, No. (%)
|
6(12.0)
|
15(32.6)
|
0.015
|
<124, No. (%)
|
40(80.0)
|
28(60.9)
|
|
>558, No. (%)
|
4(8.0)
|
3(6.5)
|
|
Endpoint
|
|
|
0.054
|
124-558, No. (%)
|
8(25.8)
|
20(51.3)
|
0.031
|
<124, No. (%)
|
18(58.1)
|
12(30.8)
|
|
>558, No. (%)
|
5(16.1)
|
7(17.9)
|
|
At the end of our study, there were a total of 15 patients lost to follow-up, of which 12 were in the PTX group and 3 were in the RFA group. And 12 patients died in total during the follow-up period. The all-cause mortality of the PTX group and the RFA group were 14.6% (6/41) and 13.6% (6/44), respectively. According to the Kaplan-Meier survival curve, there was no statistical difference in long-term survival rates in the two groups(P=0.902) (Figure 1). A total of 10 patients relapsed, and the recurrence rates were 9.8% (4/41) and 13.6% (6/44) in the PTX and RFA group(P=0.579). There was no significant difference in the cumulative response rate between these two groups (P=0.141) (Figure 2).
Secondary outcomes: iPTH concentrations in both groups decreased sharply from baseline after therapy. Mean iPTH concentrations in PTX group and RFA group immediately after treatment were 82.30±163.21 pg/mL and 279.96±306.57 pg/mL, respectively (P<0.001). iPTH levels reached a small peak in 6 months in both groups, and there was no significant difference in the trend of these over time between the two groups from discharge to 24 months(P>0.05) (Figure 3a). The postoperative serum calcium and phosphorus concentrations of the two groups were significantly lower than the baseline. The average calcium concentration of the RFA group at discharge was the lowest value of 2.04±0.33mmol/L, while it dropped to the lowest of 2.01±0.38mmol/L at 1 month in the PTX group. The lowest phosphorus concentrations of the PTX and RFA groups were in 1 month, and were 1.00±0.23mmol/L and 1.16±0.40mmol/L in that order. However, the comparison of the mean calcium and phosphorus concentrations at different follow-up times was no statistical difference between the two groups (P>0.05) (Figure 3b,3c).
Adverse events and complications
Hoarseness occurred in 4 and 6 cases respectively in the PTX and RFA groups(P=0.385). And hematoma occurred in 3 and 1 cases in the PTX and RFA groups(P=0.368). The incidence of hypocalcemia in the RFA group was 55.3%, but there was no significant difference compared with the PTX group, which was 43.1% (P=0.228). SH occurred in 10 patients, including 4 in the PTX group, and 6 in the RFA group(P=0.421). The incidence fever or infection in the PTX group was obviously more than of the others(P<0.001). There was a significant difference in postoperative CRP levels between the two groups (P=0.015). (Table 3)
Table 3. Comparison of safe between PTX and RFA group
Parameters
|
PTX Group(n=53)
|
RFA group(n=47)
|
P value
|
Total hospital stay(days)
|
15.49±8.57
|
11.60±4.52
|
0.006
|
Postoperative hospital stay(days)
|
7.85±5.92
|
4.53±2.98
|
0.001
|
CRP after treatment
|
34.20(19.00-88.55)
|
24.00(6.50-38.40)
|
0.015
|
Complications
|
|
|
|
Hoarseness, No.(%)
|
4(7.5)
|
6(12.8)
|
0.385
|
Hematoma, No.(%)
|
3(5.7)
|
1(2.1)
|
0.368
|
Fever/infection, No.(%)
|
21(39.6)
|
3(6.4)
|
<0.001
|
Hypocalcemia, No.(%)
|
22(43.1)
|
26(55.3)
|
0.228
|
Severe hypocalcemia, No.(%)
|
4(7.8)
|
6(12.8)
|
0.421
|
Hypocalcemia<2.0mmol/L; Severe hypocalcemia<1.8mmol/L
Compared with the PTX group, the total hospital stays in the RFA group (15.49±8.57 days vs 11.60±4.52 days, P=0.006) and postoperative hospital stays (7.85±5.92 days vs 4.53±2.98 days, P=0.001) were significantly less. Furthermore, the RFA group was analyzed by subgroups. There was no statistical difference between the single-session RFA group and the two-session RFA group in terms of total and postoperative hospitalization time (10.73±4.17 days vs 12.67±4.14 days, P=0.147; 4.62±3.70 Days vs 4.43±1.83 days, P=0.834). However, in terms of operating expenses, the single-session ablation of 1447.70±41.88$ was less than parathyroidectomy of 1633.85±258.84$, but the cost of two-session ablation was 2820.25±1.99$.
Risk factors of hypocalcemia
Univariate analysis showed that lower age(P=0.030) and baseline serum calcium concentrations(P=0.009), higher baseline iPTH levels(P=0.019), greater reductions of iPTH levels in D1(P=0.006), higher bALP levels(P<0.001) were associated with a higher risk of hypocalcemia (Table 4). When the cut-off point of bALP level was set as 115U/L, the area under the ROC curve was 0.762, the sensitivity was 80.0%, and the specificity was 63.6% (Figure 4). The results of binary logistic regression analysis showed that OR=1.033 (1.013-1.052), for every increase of bALP levels(1U/L), the risk of hypocalcemia increased by 3.3%.
Table 4. ROC curves in patients with hypocalcemia
Variables
|
Cut-off
|
AUC
|
Sensitivity
|
Specificity
|
P value
|
Baseline iPTH(pg/mL)
|
1210
|
0.638
|
0.833
|
0.36
|
0.019
|
Reduction of iPTH(pg/mL)
|
894.9
|
0.667
|
0.979
|
0.304
|
0.006
|
Serum calcium(mmol/L)
|
2.61
|
0.654
|
0.420
|
0.896
|
0.009
|
BALP(U/L)*
|
115
|
0.762
|
0.800
|
0.636
|
<0.001
|
Age(years)
|
47.5
|
0.627
|
0.700
|
0.542
|
0.030
|
AUC* > 0.7.