末期腎臟病病人高血磷之飲食對策

Abstract

背景：末期腎臟病病人是心血管疾病的高危險族群，並都過早死亡。新證據表明，高磷血症、纖維母細胞生長因子23 (fibroblast growth factor 23, FGF23) 過量、繼發性副甲狀腺功能亢進和維生素D缺乏等在內的礦物質代謝異常，導致該族群的不良後果。膳食磷限制是改善礦物質與骨骼代謝異常的首選方法，但其對FGF23的影響未明。針對患有高磷血症之血液透析病人，目前的臨床指引建議以磷與蛋白質比率 (phosphate-to-protein ratio, PPR) 當作工具來慎選食物，食物的PPR值應為10-12 mg/g，以減少膳食磷、同時提供足夠蛋白質攝取量。營養成份資料庫中PPR的準確性，及每日適當膳食磷攝取量目前均未明。本論文的目的是 (1)探討我國營養成份資料庫估算PPR的準確性，(2)藉由系統性文獻回顧及統合分析評估飲食對FGF23降低效果，以及 (3)研究血液透析病人低磷飲食對FGF23的影響，及每日適當膳食磷攝取量。 方法：(1)將從醫院菜單中選取20種烹飪菜餚，探討營養成份資料庫估算PPR的準確性。(2)系統性搜索四個醫學資料庫(Medline，PubMed，Embase和Cochrane Library)，收集文獻鎖定低磷飲食對FGF23降低作用的隨機臨床試驗，並執行隨機效應模型的統合分析。(3)在一項隨機交叉試驗中，招募患有末期腎臟病和高磷血症的病人，隨機分派為兩組，接受為期兩天的極低磷餐(PPR 8 mg/g，磷含量低於600毫克/天)，或低磷餐(PPR 10 mg/g，磷含量低於800毫克/天)。主要探討結果為攝取兩種低磷餐之後intact FGF23值變化的差異。次要結果包括血磷值、副甲狀腺及C-terminal FGF23 值變化。 結果：(1)不同研究食物項目間營養素的差值，以營養成份資料庫估計值減掉測量值計算，明顯不同。與每100公克食物所含PPR的測量值相比，估計值會明顯高估(平均差異值 ± 標準差，3.5 ± 12.2 mg/g，P = 0.008). (2)系統性文獻回顧顯示過去有5篇隨機對照試驗探討慢性腎臟病患者，但都沒有納入透析患者。統合分析表示，相較於攝取磷含量較高的餐點，低磷餐的攝取傾向於降低FGF23值(標準化平均差異值 -0.74, 95% CI -1.54 to 0.07, P = 0.07). (3)共有35名受試者參加研究，29名完成試驗。與基準值比較，兩種低磷餐均能降低intact FGF23，血磷及副甲狀腺，卻未能改變C-terminal FGF23值。相較於低磷餐，極低磷餐對intact FGF23，副甲狀腺或C-terminal FGF23 值並沒有額外降低作用，但能額外降低血磷值 (平均差異值, 0.6 mg/dL; 95% CI, 0.2 to 1.0; P = 0.002). 結論: 在使用營養成份資料庫評估PPR以準備醫院飲食時應謹慎行事。限磷飲食可降低慢性腎臟病患者的FGF23值。對於血液透析患者，極低磷餐對FGF23值降低沒有額外的益處，但提供更大的降磷作用。

Background: Patients with end-stage renal disease (ESRD) are at an increased risk for developing cardiovascular disease, and premature death. Emerging evidence suggests that disordered mineral metabolism including hyperphosphatemia, fibroblast growth factor 23 (FGF23) excess, secondary hyperparathyroidism, and vitamin D deficiency contribute to the high rates of adverse outcomes. Dietary phosphate restriction is a preferred way to ameliorate altered mineral metabolism, but yields conflicting results on FGF23. For hemodialysis patients who have hyperphosphatemia, current guidelines recommend the metric phosphate-to-protein ratio (PPR) to select foods with a PPR value of 10 – 12 mg/g, fulfilling low phosphorus purpose while ensuring adequate protein intake. The accuracy of PPR in nutrient database and the optimal amount of dietary phosphate restriction are undefined. The objectives of this dissertation are (1) to examine the accuracy of nutrient database estimating PPR, (2) to assess diet-mediated FGF23-lowering effect using a systematic review and meta-analysis, and (3) to investigate the effect of low-phosphate diet on FGF23, and assess the optimal amount of dietary phosphate restriction in hemodialysis patients. Methods: (1) Accuracy of nutrient database regarding PPR in 20 cooked dishes selected from the hospital menu was determined. (2) Four database (Medline, PubMed, Embase, and Cochrane Library) were systematically searched for randomized clinical trials (RCTs) that examined FGF23-lowering effects of low-phosphate diets to conduct a random-effects meta-analysis. (3) In a randomized crossover trial, adults with ESRD and hyperphosphatemia were randomly assigned to receive a very-low-phosphate diet (PPR 8 mg/g, equivalent to phosphate content less than 600 mg/d), or a low-phosphate diet (PPR 10 mg/g, equivalent to phosphate content less than 800 mg/d) for two days. The primary outcome was mean difference in change-from-baseline intact FGF23 level between intervention groups. Secondary outcomes included changes in serum phosphate, intact parathyroid hormone (PTH), and C-terminal FGF23 level. Results: (1) There is a substantial variation in differences of nutrients (estimated values based on nutrient database – measured values) among study foods. Comparisons of estimated values with measured values for every 100 g of tested foods revealed a significant overestimation for the PPR (mean difference ± SD, 3.5 ± 12.2 mg/g, P = 0.008). (2) In the systematic review, five RCTs of patients with chronic kidney disease (CKD) were identified and no study enrolled dialysis patients. The meta-analysis revealed that lower phosphate diets tended to reduce FGF23 levels, compared with higher phosphate diets (standardized mean difference, -0.74; 95% CI, -1.54 to 0.07; P = 0.07). (3) A total of 35 patients consumed study diets and 29 completed the trial. Both low-phosphate diets significantly decreased intact FGF23, phosphate, and intact PTH levels relative to baseline values, but no change in C-terminal FGF23 level. Relative to the low-phosphate diet, the very-low-phosphate diet had no significant effect on the level of intact FGF23, intact PTH, or C-terminal FGF23 but lowered the serum phosphate level (mean difference, 0.6 mg/dL; 95% CI, 0.2 to 1.0; P = 0.002). Conclusions: Caution should be exercised in estimating the PPR using a nutrient database to prepare hospital diets. Dietary phosphate restriction may reduce FGF23 levels in CKD patients. For hemodialysis patients, the very-low-phosphate diet offered no additional benefit for FGF23 reduction but provided a greater phosphate-lowering effect.