Introduction

Secondary hyperparathyroidism (SHPT) is a major complication in hemodialysis patients, which is closely associated with their survival [1, 2]. In 2006, Kidney Disease/Improving Global Outcomes (KDIGO) presented new concepts of renal osteodystrophy to cover chronic kidney disease-mineral and bone disorders (CKD-MBD) [3], and the concepts were later presented as clinical practice guidelines [4]. In Japan, guidelines on CKD-MBD were concurrently presented in 2009 [5] and revised later. Although these guidelines recommend target values for management of bone mineral metabolism, there is no standard on parathyroid hormone (PTH) values. Hence, target values for PTH vary widely among guidelines [4, 5].

Various symptoms in patients with advanced SHPT have been demonstrated to be alleviated after parathyroidectomy (PTx) [6–9]. Their prognosis after PTx is shown to be better than that of patients received conservative treatments with active vitamin D receptor activators (VDRAs) [10–14]. However, there are few reports on the relationship between PTH levels and the prognosis after PTx [15]. In this study, we retrospectively evaluated the prognosis of patients after PTx by dividing them into two groups, according to the median value of individual mean values of the postoperative PTH level.

Subjects and methods

Subjects

Three hundred and three patients on hemodialysis (3 times a week, 4–6 h per session), who had no history of either kidney transplantation or treatments with cinacalcet, underwent PTx without autotransplantation from July 1998 to the end of 2011. Among these 303 patients, 295 patients with the follow-up data were enrolled in this study. The observation period was 5.00 ± 0.28 years (mean ± standard deviation, SD), and the maximum observation period was 12 years after PTx. Based on the data obtained every 4–8 weeks during the observation period, the mean value of intact PTH (iPTH) was calculated to be 65.0 ± 104.4 pg/mL, and the mean number of measurements was 47.7 ± 34.8/patient.

Patients were allocated to one of the two groups, the low PTH (iPTH < 16.6 pg/mL) (n = 148) and high PTH (PTH ≥ 16.6 pg/mL) (n = 147) groups, using the median iPTH value of 16.6 pg/mL. Then, their background at PTx, postoperative PTH profile, mineral parameters, magnitude of uremia [anemia, blood urea nitrogen (BUN), creatinine, lipids, and C-reactive protein (CRP)], and VDRAs on the prognosis were examined. Continuous variables were dichotomized at the median value and used as prognostic factors. The primary endpoint was cardiovascular death and the secondary endpoint was overall mortality. Cardiovascular death included death due to cerebral infarction/hemorrhage, subarachnoid hemorrhage, myocardial infarction, and heart failure, as well as sudden death. PTx was performed under general anesthesia, and visible parathyroid glands were extirpated as completely as possible. The number of extirpated glands was 4 or more in 221 patients, and 3 or fewer in 74 patients. Based on laboratory findings obtained every 2 weeks, doses of VDRAs, phosphate binders, and erythropoietin stimulating agents were determined, using target ranges as below; 3.5 ≤ phosphorus < 6.0 mg/dL, 8.0 ≤ calcium < 9.5 mg/dL, and 10.0 ≤ hemoglobin < 12.0 g/dL.

Data collection

Blood samples were collected at the beginning of the first hemodialysis of the week. Mean values were based on the data obtained every 4 weeks for serum levels of calcium, phosphorus, albumin, alkaline phosphatase (ALP), hemoglobin, BUN, creatinine, and CRP. Mean values were calculated with the data obtained every 4–8 weeks for iPTH and every 1–6 months for total cholesterol (T-CHO), high-density lipoprotein cholesterol (HDL-C), and triglycerides, respectively. The iPTH level was analyzed by an immunoradiometric assay from the start of the study to March 2003, and the method was changed to an electrochemiluminescence immunoassay from April 2003. As a routine screening examination before surgery, 12-lead electrocardiogram (ECG) was checked, and a cardiologist diagnosed abnormal findings on the ECG, such as left ventricular hypertrophy, abnormal Q waves, inverted T waves, nonspecific and specific ST changes, arrhythmias, and conduction disturbances.

Statistical analysis

All results were expressed as the mean ± SD. The unpaired t test was used for analysis of continuous variables, while the Fisher’s exact test was used for categorical data. The Kaplan–Meier method was employed to calculate survival rates, and the log-rank test was performed to assess the significance of differences in survival. The multivariate analysis of prognostic factors was performed using the Cox proportional hazard model. All tests were two-sided, and P < 0.05 was considered significant. Statistical analyses were performed with JMP Version 7.0 (SAS Institute Inc., Cary, NC, USA).

The Ethical Committee for Human Research of Tojinkai Hospital approved this study, and the study was performed in accordance with the Principles of the Declaration of Helsinki.

Results

Patient profiles at PTx are shown in Table 1. In the low PTH group, the incidence of cardiovascular complications and the frequency of abnormal ECG were significantly lower than those in the high PTH group. There were no differences between the two groups with regard to age, sex, dialysis history, diabetes, body mass index (BMI), Kt/v, antihypertensive therapy, and blood pressure at the time of surgery. In the low PTH group, serum calcium and albumin levels were significantly higher at the time of surgery, although there were no significant differences in other variables, such as serum iPTH and phosphorus values. 6 days after surgery, significantly lower iPTH levels were observed in the low PTH group (Table 2).

Table 1 Patients’ profiles at parathyroidectomy
Full size table
Table 2 Laboratory findings at parathyroidectomy
Full size table

The data obtained after PTx are shown in Table 3. The low PTH group showed significantly lower serum calcium, phosphorus, iPTH, and T-CHO levels. The ALP level tended to be lower in the low PTH group, although the effect was not significant. There were no differences in serum levels of albumin, hemoglobin, creatinine, CRP, HDL-C, and triglycerides. No difference was found in the duration of VDRAs administration.

Table 3 Laboratory findings during the observation period after parathyroidectomy and the dosing period of VDRAs
Full size table

The rate of extirpation of more than 4 glands was 58.5 % in the high PTH group, which was significantly lower than that in the low PTH group (92.6 %, P < 0.001). Moreover, the mean value of PTH measured during the observation period was strongly correlated with the iPTH value measured 6 days after PTx (Spearman ρ = 0.9007, P < 0.0001, y = 0.4725x + 30.395, R 2 = 0.51798) (Fig. 1).

Fig. 1
figure 1

Correlation between the serum level of iPTH measured 6 days after PTx (iPTH6) and the mean value of iPTH after surgery

Full size image

No patient died in the immediate postoperative period. During the observation period, 43 deaths were reported, comprising 17 deaths (11.5 %) in the low PTH group and 26 deaths (17.7 %) in the high PTH group (P = 0.1313). There were 17 cardiovascular deaths, including 4 cases (2.7 %) in the low PTH group and 13 cases (8.8 %) in the high PTH group (P = 0.0236). Overall survival did not differ significantly between the two groups (P = 0.1500). However, cardiovascular death-free survival rates in the low and high PTH groups were 94.4 and 71.6 %, respectively (P = 0.0256) (Fig. 2). In a multivariate adjusted model (n = 295, event = 17) for sex, age, history of cardiovascular disease (CVD), dialysis duration, iPTH, albumin, calcium, phosphorus, and ALP, major factors in the CVD mortality risk were age ≥ 59 years (hazard ratio, 2.654; 95 % CI 0.964–7.306; P = 0.0589), history of CVD (hazard ratio, 3.377; 95 % CI 1.112–10.254; P = 0.0317) and iPTH ≥ 16.6 pg/mL (hazard ratio, 3.132; 95 % CI 1.015–9.664; P = 0.0470) (Fig. 3).

Fig. 2
figure 2

Cumulative survival rates. a Overall survival and b cardiovascular disease (CVD) death-free survival. Thick and thin lines represent survival rates in the low and high PTH groups, respectively. OAS overall survival

Full size image
Fig. 3
figure 3

Cox regression multivariate analyses for cardiovascular death using factors of P < 0.20, after univariate analyses using sex, age, history of cardiovascular disease (CVD), dialysis duration (DD), iPTH, albumin, calcium, phosphorus, and alkaline phosphatase, without a missing value (n = 295, 17 deaths). Calcium values were adjusted by Payne’s equation

Full size image

Discussion

The influence of serum concentrations of phosphorus, calcium and PTH on mortality has been discussed independently. Despite their interdependency, disordered mineral metabolism has never been assessed in the context of all three parameters. Although high levels of phosphorus and calcium in the high PTH group had been supposed to increase cardiovascular mortality, we found that postoperative serum concentrations of phosphorus and calcium were not significant, but iPTH median ≥ 16.6 pg/mL was a significant factor in cardiovascular death in the Cox proportional hazard model.

Amelioration of hypertension [6] and attenuation of cardiac hypertrophy after PTx have been reported [16]. Since fibroblast growth factor 23 (FGF23) facilitates hypertrophy of cardiomyocytes [17], the beneficial outcome found in the present study may be partly attributed to recovery of cardiac function caused by the decrease in the FGF23 level after PTx [18, 19]. The coronary artery calcification is also alleviated or stable after PTx [20, 21]. Fetuin-A is an inhibitor of extraosseous calcification [22], and an inverse correlation between fetuin-A and iPTH levels after PTx is reported [23]. On the other hand, the circulating level of fetuin-mineral complex, which reflects extraosseous calcification stress, is shown to decrease after either PTx or treatments with cinacalcet [24]. These may all contribute to suppression of cardiovascular death associated with cardiac dysfunction and vascular calcification of the coronary artery. A study on the relationship between the vascular system and PTH showed that the number of vascular endothelial precursor cells increased in peripheral blood along with a decrease in the PTH level [25]. In another study, PTx improved coronary flow reserve in primary hyperparathyroidism [26]. These organic and functional improvements of the cardiovascular system following PTx may be involved in the postoperative reduction of cardiovascular mortality. In addition to the effect of PTx on the cardiovascular system, improvements in muscle strength, sleep, and cognitive ability after surgery [7–10] seem to improve activity of daily living and thus contribute to the prognosis.

Although suppression of the PTH level by PTx increases bone density, there is a risk of fracture associated with adynamic bone disease due to excess suppression of the PTH level. However, Rudser et al. found a clear decrease in the fracture rate after PTx, and they suggested that an increase in muscle strength and an improvement of activity of daily living contributed to prevention of fracture after PTx [27].

In 1993, we changed our surgical methods from subtotal PTx to total PTx with autotransplantation. However, the recurrence of hyperparathyroidism (iPTH ≥ 200 pg/mL) accompanied by high phosphorus levels was frequent (unpublished data). Thus, we changed again the surgical procedure to total PTx without autotransplantation in 1998. Complete excision of parathyroid glands is difficult, because of their variation among individuals, such as abnormal locations and existence of accessory glands [28]. Even in patients whom we deemed to have completed total PTx without autotransplantation, a minute amount of PTH could still be measured. This may be caused by another source of parathyroid hormone besides parathyroid glands, as shown in a mouse study [29]. For these reasons, total PTx (excision of more than 4 glands) without autotransplantation [30, 31] is judged to be a practical method for treating SHPT.

The ratio of extirpation of more than 4 glands was 58.5 % in the high PTH group, which was significantly lower than that in the low PTH group (92.6 %, P < 0.001), and the iPTH day 6 and mean postoperative iPTH showed a strong correlation (Spearman ρ = 0.9007, P < 0.0001, y = 0.4725x + 30.395, R 2 = 0.51798). Therefore, the postoperative iPTH value seems to be sufficient to assess complete resection of the glands by PTx.

There are some limitations in the present study. First, the sample size was small, and the study was carried out retrospectively with a relatively short observation period of about 5 years. Second, this study did not clarify medications for patients suffering from cardiovascular disease and factors contributing to non-fatal cardiovascular disease. Third, VDRAs [32] and phosphate binders [33] could affect the prognosis, although they were not fully investigated. The low PTH group received very low doses of oral VDRAs (alfacalcidol 0.75–1.75 µg/week), while therapy in patients with persistently high PTH levels or postoperative recurrence (iPTH ≥ 200 pg/mL) was switched from oral administration of VDRAs to intravenous injections of VDRAs, concurrent administration of cinacalcet, or re-operation. Regarding phosphate binders, management of serum phosphorus has changed with release of non-calcium-containing binders (sevelamer hydrochloride in 2004 and lanthanum carbonate in 2009). Non-calcium-containing phosphate binders might have exerted an influence on the prognosis.

Despite various confounding factors, this study showed that the low PTH group had fewer cardiovascular deaths than the high PTH group. Further investigations on a large scale are required to clarify how the PTH level affects the prognosis after PTx and to define an optimal range of PTH.

Conclusion

The findings of the present study suggest the possibility that maintaining low PTH levels after PTx reduces cardiovascular mortality and improves the prognosis. Total PTx (more than 4 glands) without autotransplantation seems to be one of the options for managing severe SHPT.