Baseline characteristics

The study population consisted of 111 men with prior oil injections, where the severity of the disease was assessed by stratifying the men according to the degree of calcium imbalance: Men with normocalcemia, or men with manifest hypercalcemia (ionized calcium > 1.32 mmol/L and/or albumin corrected calcium > 2.51 mmol/L) (Table 1). The mean age of all men was 38.4 years (SD 7.9), with a mean BMI of 28.4 kg/m² (SD 3.4). There were no differences in the two groups in age (p = 0.147), height (p = 0.654), weight (p = 0.578), BMI (p = 0.420), smoking (0.188), or alcohol use (0.375). The men had injected an average volume of 951 mL (SD 876) of oil, with hypercalcemic men injecting more than men with normocalcemia (1189 mL vs. 852 mL, p = 0.091). The average time since oil injection was 12.1 years (SD 3.6) across all men, also slightly longer in hypercalcemic men than in normocalcemic men (12.8 years vs. 11.4 years, p = 0.077). As men were stratified according to calcium concentrations, total and ionized calcium was higher in hypercalcemic men compared to normocalcemic men (2.69 mmol/L vs. 2.36 mmol/L, and 1.46 mmol/L vs. 1.24 mmol/L, respectively). In response, PTH concentrations were significantly lower in hypercalcemic men (1.0 pmol/L vs. 2.6 pmol/L, p < 0.001). Average 1,25(OH)2D3 concentrations were also higher in hypercalcemic men (154 pmol/L vs. 116 pmol/L, p = 0.005), while the mean 25(OH)D level was slightly higher as well (54 nmol/L vs. 46 nmol/L, p = 0.081). Magnesium (p = 0.446), phosphate (p = 0.762), sodium (p = 0.088), and potassium (p = 0.298) concentrations did not differ significantly between groups. Iron levels were lower in hypercalcemic men compared to normocalcemic men (16.3 µmol/L vs. 19.2 µmol/L, p = 0.041). Hemoglobin levels were also significantly lower in the hypercalcemic group (9.3 mmol/L vs. 9.9 mmol/L, p = 0.012).

Associations between oil injection and average serum mineral concentrations

The severity of granuloma disease was also assessed by stratifying men according to the self-reported amount of oil injected (< 500 mL, 500-1,000 mL, 1,000–2,000 mL, and > 2,000 mL), as shown in Fig. 2. Serum concentrations of calcium and PTH were correlated with the amount of oil injection. Men who had injected > 2,000 mL had higher total calcium (2.57 mmol/L vs. 2.40 mmol/L, p = 0.029, Fig. 2A) and ionized calcium (1.36 mmol/L vs. 1.25 mmol/L, p < 0.001, Fig. 2B) compared to men who had injected < 500 mL. Accordingly, serum PTH concentrations were lower (1.7 pmol/L vs. 3.2 pmol/L, p < 0.001, Fig. 2C), but also when compared to men who had injected 500-1,000 mL (1.7 pmol/L vs. 2.2 pmol/L, p = 0.008) and 1,000–2,000 mL (1.7 pmol/L vs. 2.4 pmol/L, p = 0.031). While there was a slight trend of lower 25(OH)D, and higher 1,25(OH)2D3 concentrations with more amount of oil injected, the differences were not significant (p = 0.18 and p = 0.76, respectively, Fig. 2D and E). However, the men who had injected > 2,000 mL had lower serum magnesium concentrations compared to the other groups, 0.75 mmol/L vs. 0.79 mmol/L, 0.81 mmol/L, and 0.82 mmol/L, respectively (p < 0.001, p < 0.001, and p = 0.042). They also had higher sodium concentrations compared to < 500 mL (141 mmol/L vs. 139 mmol/L, p = 0.048, Fig. 2H). No differences were found in phosphate (p = 0.44), potassium (p = 0.86), and iron concentrations (p = 0.12) (Fig. 2G and I, and 2J).

Fig. 2

Association between the amount of oil injected and concentrations of serum calcium, PTH, 25(OH)D, 1,25(OH)2D3, and serum minerals. Serum concentrations of minerals grouped by self-reported amount of oil injected. (A) Total calcium, (B) Ionized calcium, (C) PTH, (D) 25(OH)D, (E) 1,25(OH)2D3, (F) Magnesium, (G) Phosphate, (H) Sodium, (I) Potassium, and (J) Iron. Green crosses represent each individual and black lines represent the mean with standard deviation. P-value for differences between the groups was assessed by a one-way ANOVA test and post-hoc pairwise comparisons adjusted with Dunnett’s test. Abbreviations: PTH: parathyroid hormone; 25(OH)D: 25-hydroxyvitamin D; 1,25(OH)2D3: 1,25-dihydroxyvitamin D

Changes in total calcium, PTH, 25(OH)D, and 1,25(OH)2D3

When stratified according to baseline calcium concentrations, men with hypercalcemia had a decrease in total calcium over time, while men with normocalcemia remained close to baseline concentrations for up to 48 months (Fig. 3A). Men with hypercalcemia dropped from a baseline concentration of total calcium at 2.69 mmol/L (SD 0.30) to 2.46 mmol/L (SD 0.15) after 48 months, while men with normocalcemia had slight changes from 2.36 mmol/L (SD 0.08) to 2.38 mmol/L (SD 0.10). When normalizing concentrations to baseline, and comparing the groups at 48 months, there was a significant difference (p < 0.001) in the values (Fig. 3A). The same changes were observed in albumin corrected calcium (Fig. 3C). The opposite reaction was seen in serum PTH, where men with hypercalcemia had a decrease in PTH from 1.0 pmol/L (SD 0.7) at baseline, to 2.1 pmol/L (SD 2) after 48 months, with a significant difference (p < 0.001) when compared to men with normocalcemia (Fig. 3D). Also, concentrations of 25(OH)D showed a decrease in men with hypercalcemia, even though the values were in the normal range at baseline, reflecting the restriction in vitamin D intake and sun exposure (Fig. 3E). 1,25(OH)2D3 decreased from 154 pmol/L (SD 66) at baseline, to 97 pmol/L (SD 48) in men with hypercalcemia after 48 months (p = 0.042) (Fig. 3F).

Fig. 3

Measured concentrations of total calcium, PTH, 25(OH)D, and 1,25(OH)2D3 for hypercalcemic and normocalcemic men from baseline with follow-up to 48 months. (A-D) Normalized plots for serum concentrations of total calcium, PTH, 25(OH)D, and 1,25(OH)2D3 in patients grouped according to calcium status at baseline. Data from baseline, and after 1, 3, 6, 12, 19, 24, 36, and 48 months are shown. Differences between groups at 48 months were compared by a crude comparison of means (t-test). * p-value < 0.05

Changes in ionized calcium, magnesium, phosphate, sodium, potassium, and iron

Ionized calcium showed a similar trend to total calcium, as men with hypercalcemia dropped and ended on different (p < 0.001) normalized values compared to men with normocalcemia (Fig. 4A). Of the 79 men who were normocalcemic, only 5 (6%) developed hypercalcemia (ionized ca. > 1.32 mmol/L) after 36 months, and only 2 (2.5%) after 48 months (data not shown). While men who were normocalcemic at baseline did not experience a change in serum phosphate through the follow-up period, men with hypercalcemia saw a rise from 1.01 mmol/L (SD 0.41) at baseline, to 1.10 mmol/L (SD 0.46) in men with hypercalcemia after 48 months (p = 0.042 when compared to men with normocalcemia). Serum iron concentrations had a transient increase to their highest point at 17.1 µmol/L after 18 months in men with hypercalcemia but decreased back to baseline levels, and no changes were seen in ferritin concentrations (data not shown). Concentrations of magnesium (Fig. 4B), sodium (Fig. 4D), and potassium (Fig. 4E) did not change or differ between the two groups over time. Kidney function improved slightly in men with hypercalcemia from a baseline eGFR of 69 mL/min/1.73 m2 (SD 21) to 73 mL/min/1.73 m2 (SD 21) but with no difference compared to the normocalcemic group for either eGFR or serum creatinine (data not shown).

Fig. 4

Measured concentrations of ionized calcium, magnesium, phosphate, sodium, potassium, and iron for hypercalcemic and normocalcemic men from baseline with follow-up to 48 months. (A-F) Normalized plots for serum concentrations of ionized calcium, magnesium, phosphate, sodium, potassium, and iron in patients grouped according to calcium status at baseline. Data from baseline, and after 3, 6, 12, 19, 24, 36, and 48 months are shown. Differences between groups at 48 months were compared by a crude comparison of means (t-test). * p-value < 0.05

Changes in serum minerals after prednisolone treatment

A subgroup of 21 men were treated with prednisolone at varying lengths during the follow-up. The average dose of treatment was 23 mg/daily (SD 15) at baseline, decreasing to 17.5 mg/daily (SD 11.9) after three months, 15 mg/daily (SD 7.5) after six months, 8.75 mg/daily (SD 5) after twelve months, 5 mg/daily (SD 5) after 18 months and 24 months, before increasing slightly to 7.5 mg/daily (SD 5) after 36 and 48 months. Figure 5 shows serum concentrations of ionized calcium, magnesium, phosphate, sodium, potassium, and iron normalized to baseline for patients treated with prednisolone from baseline and up to 48 months. Serum ionized calcium concentrations gradually dropped in the follow-up period (Fig. 5A). Baseline average calcium ion levels were measured at 1.52 mmol/L (SD 0.20), which decreased to 1.44 mmol/L (SD 0.22) after three months (p = 0.240), 1.40 mmol/L (SD 0.16) after six months (p = 0.042), and 1.38 mmol/L (SDs 0.13 and 0.15) after 12 and 18 months (p = 0.028 and p = 0.036, respectively). It further dropped to 1.34 mmol/L (SD 0.13) after 24 months (p = 0.011), before normalizing at 1.32 mmol/L (SD 0.14) after 36 months (p = 0.035) and ending at 1.27 mmol/L (SD 0.06) after 48 months (p < 0.01). Serum magnesium concentrations also followed the same trend (Fig. 5B) starting with baseline average concentration at 0.81 mmol/L (SD 0.11), before dropping to 0.76 (SD 0.11) and 0.77 mmol/L (SD 0.11) after three and six months (p = 0.030 and p = 0.045). It further dropped to 0.75 mmol/L (SD 0.15) after 18 months (p = 0.042), 0.76 mmol/L (SD 0.13) after 24 months (p = 0.025), and 0.75 mmol/L (SD 0.09) after 36 months (p = 0.027) and ended on 0.77 mmol/L after 48 months (p = 0.251). There was a small drop in phosphate concentration after six months, decreasing from baseline average concentration of 1.05 mmol/L (SD 0.26) to 0.87 mmol/L (SD 0.22), but later increased again over time to baseline levels (Fig. 5C). Finally, there was an initial increase in iron concentration, where it rose from baseline average concentrations at 14.9 µmol/L (SD 6.3) to 17.6 µmol/L (SD 6.8) after 6 months (p = 0.039) but later normalized close to baseline again over time (Fig. 5F). No significant changes were observed in sodium and potassium concentrations (Fig. 5D and E).

Fig. 5

Measured concentrations of ionized calcium, magnesium, phosphate, sodium, potassium, and iron for patients treated with prednisolone from baseline with follow-up to 48 months. (A-F) Normalized plots for serum concentrations of ionized calcium, magnesium, phosphate, sodium, potassium, and iron in patients who have been treated with prednisolone. Data from baseline, and after 3, 6, 12, 18, 24, 36, and 12 months are shown. Differences between groups were compared by a linear mixed model for repeated measurements and Dunnett’s test to adjust for multiple comparisons. * p-value < 0.05