‘PERSPECTIVES IN PRACTICE Potential renal acid load of foods and its influence on urine pH THOMAS REMER, PhD; FRIEDRICH MANZ, MD ‘The purpose of this study was to calculate the potential renal acid load (PRAL) of selected, frequently consumed foods. A physiologically based calculation model was recently vali- dated to yicld an appropriate estimate of renal net acid exeretion (NAB); the model depends primarily on nutrient Intake data. When nutrient data from actual food composition tables were used, the calculation model yielded PRAL values that ranged from an average maximum of 23,6 mEq/100 g for certain hard cheeses aver 0 mEq/100 g for fats and cils to an average minimum of approximately -8 mBq/100 g for fruits and fruit juices and vegetables. By means of these PRA, data (summed according to the amounts of foods and beverages ‘consumed daily and by an estimate of excretion of organic acids [based on body sizel), the daily NAE can be calculated. ‘This calculation methodology, primarily based on PRAT, allows an appropriate prediction of the effocts of diet on the acidity of urine. For practical applicability in dietetic pre- vention of recurrent urolithiasis or in other flelds of dietetics, the additionally determined correlation (7.83; P<.001) between NAB and urine pH can be used to ascertain NAB. target values for a desired urine pH modification, Am Diet Assoc. 1995; 95:791-797. T. Romer is a research associate in the Working Group Nutritional Medicine and F. Manz is professor of ‘pediatrics and vice director of the Research Institute of Child Nutrition, Dortmund, Germany. “Address correspondence lo: Thomas Remer, PhD, Forschungsinstitut fur Kindererndihrung, Heinstick 11, 44225 Dortmund, Germany. rolithiasis plays a quantitatively important role among the urologic diseases. Renal hydrogen ion excretion (ie, the urine pH) is generally accepted as a urinary risk factor in most types of urinary stone disease (1). Al- though scientists have known for several decades that the ‘composition of the diat influences the urine pH (2), only recently has clear, experimental evidence been provided that it is possible to efficiently modify or adjust the urine pH by purely dietetic means (3-5). Because previous methodologic cfforts failed to provide reliable estimates of the net acid loads produced by diets (2), ‘we Lested a different, physiologically based calculation model developed for the prediction of renal net acid excrotion (NAE) from nutrient intake data (4). This calculation model, which, takes into account the mineral and protein composition of foods, the average intestinal absorption rates of the respective nutrients, sulfur metabolism, and urinary excretion of organic acids, proved to be appropriate for the prediction of NAE (4). In addition, we were able to predict and adjust the urine pH of healthy adults (5) when we used this model along with the correlation (shown in this article) between Uke analytically determined NAE and the urine pH. On the basis of these {ndings and earlier studies on protein hydrolyzates (6), syn- thetic amino acid mixtures (6), and milk formulas (7,8) — each confirming the applicability of the calculation model — it now appears justified and possible to estimate the potential renal acid load (PRAL) of foods. ‘The purpose of this study was to calculate and specify the PRAL of selected, frequently consumed foods Cquoted per 100 ‘g) and to demonstrate how Lo use Lhese data for the estimation of NAE and for prediction of the corresponding urine pH in persons consuming definite diets. In addition to the dietetic prevention of recurrent urolithiasis, our findings could be of practical relevance in the arcas of urinary tract infections, osteoporosis, and sports nutrition. MATERIAL AND METHODS Calculation Model ‘The method used for calculation of the PRAL of foods is based. on the calculation model we developed for prediction of NAE from nutrient intake data (4). The direcity determined NAE (based on urine analyses) is calculated it the conventional manner as the sum of titratable acid and ammonium minus bicarbonate (Figure 1). As is discernible from Figure 1, NAE can also be determined indirectly from the difference of the sum of the remaining important urinary anions — chloride, phosphorus, sulfate, and organic acids (nonbicarbonate an- ions) — minus the sum of the mineral cations — sodium, po- tassium, calcium, and magnesium. The amounts of these elec trolytes in urine are determined primarily by nutritional in- take. In the case of organic acids, the major determinant is body surface area or simply body weight (9). Ifgrowth, mineral losses through the skin, and transient, metabolic non-steady- state conditions are not considered (which appears to be a rational approach for healthy, nonpregnant adults under nor- JOURNAL, OF ‘THE AMERICAN DIETETIC ASSOCIATION / 791Catons Aone FIG 1. Typical urine tonogram (showing all quantitatively important urinary anions and cations) of a healthy adult consuming a protein-rich ‘S0,=sulfate; P=phosphorus; Clachloride. The directly ‘mousured net acid excretion (NAB,,,..) equals TA+NH,-HCO, Indirectly determined net acid excretion corresponds to the sum af the ‘nonbicarbonate anions minus the sum of the mineral cations (NAE pjjqej~(l+P +50 +0A)-(NavK+CarMg) Tablet ‘Average intestinal net absorption rates for minerals and protein in ‘adults consuming mxed dies and derived factors for conversion of runt iraks (illrams por 10 9 food tomieauialens (quoted ‘82 100 q Toad) estimating the food-nduced urinary excretion of the corresponding electoytes uionts" % Absorption References. Urinary Conversion factors” (mg tone tntake > mEq excretion) = 8 28 Na 00818 K 50 year 000005 co oe 28 Ca 00125 Mo 2 Ma 000068 or % 2% © oozes P os A PO, 00366. Proen 75 50 SOL 04868 10-8 "Ra sodium: K= potastiom; Ca=calcum: Mg=magresium: CI=Chiovde: P=prospnotus, PO, = paosphato; SO, sulfate ‘re veld te cervertion factors % abborpton was divided by he respective atorle weight (Na~=230, K= 99.1; Ca=401: Mg=243: Cl=355; P=31 0) fand by 100, thereby allowe the estimation ofthe utinary excreton ol the nuents (xprosted immo). Foc Ca ana Mg, theinic valence (x2), andor P the grade of casociaton ofp 74 (18), was also consicored feading foe Unt mEq), For otal poten an average content of 2.4% methionine anc 2.0% ‘Chena was assure (4) The atomic weghs of methionine (1492) an cys ‘one (1212) were Usedto estate nermetablizad meq SO, (=mmolS0, x2) ‘omoted tom absorbed amounts ¢eulureonaning amin acs alos have been calculated trom the rapresion equations for the daly mineral rake ara the coresperding unary eectoyie excretion rats (a3 Geserbod in te felerences). For calculation an avorage daly intake ofthe fespectve nutrient accorsing to tho US Racommerdie Dietary Alowance (31) {reternce ad, 250 0 years od) was ascurod (Ca= 800 mg/day; P= B00 mae Has ay "goatee cera ware en ee eraged) ——_—<————$— 1792 / JULY 1995 VOLUME 95 NUM ral living conditions), urinary excretion of electrolytes corre sponds to amounts absorbed intestinal. Thus, renal excre- tion of NAB-relevant electrolytes (ie, NAE itself) can be esti- ‘ated from nutrient intake and anthropometric data (4). For our study, the average, absorbable amounts of all rel evant nutrients (representing NAB, Without organic acids, see Figure 1) were estimated for selected foods and beverages from data on the nutrient composition of foods (per 100- cadible portion, as provided by current food tables) and from average net absorption rates taken from the literature (Table 1). Organic acid excretion, the primarily diet-ndependent component ofthe NAB, gq, isnot immediately considered for estimation of foodborné BAL, However, when estimating the [NAE of persons consuming known amounts of definite foods, daily organic acid excretion must be taken into account. As can be deduced from the equations given in the legend of Figure |, the indirect determination of NAB (and consequently ‘estimation ofthe PRAL) involves adding anions and subtract- ing cations with partly different charges and dissociation prop erties, To manage this adequately, the units of measure ofthe relevant base- and acid-forming elements (milligrams per 100 food) were converted tomilliequivalents (mEq) as described in Table 1, With respect to the metabolic conversion of organi- cally bound sulfur to sulfate, the average content of sulfur- containing arnino acids in protein was also considered. For further characterization of foods according to particu- Jarnutrient categories that appear to be major determinants of the respective acid- or base-forming potential, urinary phos- phorus excess ard alkali excess were calculated as described in the notes to Table 2. The mutrient data used for PRAL calculation were taken nearly exclusively from reference 10. ‘Additionally used sources are given inthe footnotes toTable2. Subjects ‘Twenty-four-hour urine samples from 63 volunteers aged 16 to 49 years were analyzed for parameters of acid-base status to characterize the relationship between renal NAB and urine pH. ‘The subjects were participants in studies of the Research Institute of Child Nutrition (Dortmund, Germany). Allsubjects described themselves as being healthy and none had a past medical history of renal, endocrine, metabolic, or cardiovascu- lar disease, Because of the lack of an adequately large sample of adult female volunteers who collected timed 24-hour urine samples, only males were included in the investigation, ‘Males with a high protein intake (170:+60 g/day) were drawn froma study group of young bodybuilders (16 to 29 years old) in whom the effects of dietary protein on metabolic and renal parameters were investigated (11,12). A corresponding con- trol group (11,12) consisted of adults and adolescents (>16 ‘yearsold) consumingnormalmixed diets (protein intake=98420 ‘glday). To yield urine pH data related to low NAE values, specimens of males who wero consuming. lactovegetarian diet (comparable to that described previously [4] ~~ protein con- tent=50 g/day) were also analyzed. ‘Three subjects were excluded from the final set of data because of an incomplete 24-hour urine collection (n=1) or ‘extreme ingestion of protein supplements (protein intake>300 ‘day) associated with biochemical indications ofchronic renal ‘acid stimulation (n=2). Therefore, 60 subjects were consid- ered in regression analysis. ‘Sample Collections, Analyses, and Statistics Collections of timed 24-hour urine samples and measurements of the parameters of acid-base status (urinary pH, titratable acid, ammonium, and bicarbonate) were carried out as de- seribed previously (4). The correlation between urinary pHTable2, [Nutrient* content (10) and estimated potential renal acid load (PRAL)? of 114 frequently consumed foods and beverages (related to 100-9 exible portion) Food group and ood Energy Prowin Na K = Ca MgCl," PEX® ALEX” PRALY ial ao es Beverages Bear crt x 03 2 8 1 9 3 @ of of 04 -02 Bost pale (ole, he)> % 05 5 % 4 8 2m 3% 02 oF oo 08 Boer stout botied 3 03 SB % 8 8 i # Or 03 08 -a1 (Coca-Cola 00 8 3) 64 6 8 DOs (Cocoa mad wih sem-skimmed rik 38 7 17 120 2 100 WO 37 16 37 -04 efi, infusion, S minute. 2 02 0 "2 6 2 'O of ot 14 W14 ‘ierl water (Apoinars)* o 90 # "3 9 1% 0 w oo -o4 15 -18 Mineral water (voc) o 00 1 + 4 1 8 1 oo aa 00 ot Roane’ 6 02 1 10 7 1 4 OF Of 26 Ba Tea Ian, nso ca 0 7 oO 1 1 ‘9 00 00 03 -o3 ite wine. dry 6 Ot 4 6 8 & 6 00 -or 4) “42 Fate andote Butler 7705. m5 ess Margate ms 020 m5 1 100 Ot ok 10-05 feat a9 00. 0 9 9 O 0 “0 oo 00 oo a0 ‘Suntoner seed ot e900 0 0 0 6 5 00 oo 0 Fa ea, tes % 174 7 2 1% 2% 17 10 85 $4 68 74 Hadeck 7 ie = 120 30): erring 2 168 6 M0033 2108 Bz BS TT 70 Tat Brown steamed 135238 Ce ee Frat, nuts, and ul julcow Appleluce, uneweetened eon 2 mw 7 5 6 3 09 a9 2 -22 ‘Apples, 1Svarelis.esnandskin awrepe = 47d 3 4 5 1 0 02 o2 26 -22 Aprcots 3108 2 20 1% 1 2 3 04 03 55 -a8 Bonenas 3 12 re ee er 55 Black currants 2 08 3 30 0 7 4% 15 04 Of 73-85 Creres @% 09 1 2 «13 oat 0 ks 8 Grape ice, unswostoned 4 03 7 ‘8 7 34 6 of of 43-10 Hares 50141 5 730 0 160 300 18 69 50 uy -28 wit 1 4 0 5 6 3030S 81a oman jicw 7 og 1% 07 FT 8 3 OD range ce, unsweetenec 9 05 1 15 1 8 13 98 o2 of 32 -29 Oranges at 8 10 47 10 2308 032 =a Poaches 3310 1 to 7 9 690s os 33 “28 Panu plain set 256. 2 0 s 20 490 7 125 95 136 83 oars 3 vrs, flesh and shin average 40 03 3 0 1 7 13 1 Of a2 (32-29 Preapple a 04 2 10 1 1 1 2 02 -03 26 -a7 Raisins me 24 6 100 0 43H tO BT at Stawtorroe 27 0B | 16 1 BOK A802 Waruts os 147 7 450 «94-180 980 2k72 SB Watermelon 31 OS 2 Ww 7 8 8 ‘0 02 06 21 -19 ‘Gran products Breas ye fous minac"* a ge S752 18a 4D Bread ye four™ cr re ‘road, wheat four mixed?* me 6203? 7a? e308 Bread, wheat four whole meat® te 70 © «380-2703 N8G:SASCS SA OSB Bread, write wheat zs 84 D101 DTS Corfakos 580 79 «4410 1001845060 3B 13-09 rispread, oe Ser 94 220500 AS_C«tODgI.s 748d Noodes ep aor 121180 502832008888 BSB Catfakes rled cas (Halertocken)? 355 125, 5 305 S190 GOI B11 100 S47 Rice, own ser 67 3 20 i Ho 310 20 33 83 -o9 25, Fie, wht, easy cook ae 73 % i 51 @ 15 “to 38 40 30048 Ace, wnt, easy cook, boiled 1s 28 4 [ete on ees 3 8 oa Bye ox whole a5 82 1 40 «$2 Ot Spaghet wate 3a 120, 2 20 0% % 190 25 59 ‘52 46 65 Spaghot whole mel See 194 = 130 3001208027 Wineatiut wnt, plan aa oa 3 i 15 (20 NO et 646 «(3310 ‘White fleur wholemeal! 310127 3 MO 38 120 200 BPG Legumes Beare preenFrench beans 219 0 20 49% 7 3 9 09 0s 45-8 Lentis green and brown, whol, died ver 3 we MD of MO 9800 a7 119 80 eS Pas 68 + 30 10's e's Continuad on the newt page JOURNAL OF THE AMERICAN DIETETIC ASSOCIATION / 788Sey PERSPECTIVES IN PRACTICE ‘Table 2 (cont) Nair conta (10) and estimated potental rena acid oad (PRAL}* of 114 fequenty consumed fod and beverages (elated 10 100-9 Food group and food Energy Protein Na-K © Cas Mg_s#P SS CL_=S«SO, PX ALEX PRALY Teal —Laoao a Meat and meat producta ° ~ 7 Beet. lean ony 13 203 3900-7 8) 99 oo CChckan meat ony i 20 8 %o 10 2 20 7 00 BS 7B omnes bos canned ar Bo | soak HD tag tan 3B 3B Frenkurors™ ar “95 $80 8S Hope 20? ver sausage® BO 129 «© a8) 170232290 tsakGSTB OBS 106 Unchoon meat canned 33 126 «180 MOSS BO tei? G2 BB tn Pork. ean ony ur 207 7% 30 86|'8 2 «1 «88 8B TB Frm sta an and fat 197 183 Sm 8 mo 4 82k 78 Salamit ao Ws ses e010 108 8M 94 SSS Turkey mast ony wor 213 30 0B 23 19 BOT B71 88 Veat hit joo 2170 38) wT BB.) 80 iti dtr products, end epos Bute so 35 sy uy 09 18 tT 18 ars Camember® zy 290100 DSH 0r 102 oat. Grecidertybe reduced fat 2s SS GDH]. N12 02 aaa Grosse Gouda a7 bo | «Oat 74) B49 SDT, 77 OBE Cottage cheese lan rd Grears ter sour 205 28. a0) «10 ey 8k te 1B gga chicken, whle ia 5 1401S? 12 00m tS aw Epps wht! 'e '900 1900S BA gs yok 399161 3 121915 0D 78 83a Froth choose (Quat)” i225 Ee a Furtat sotcheese aig ‘ee = sao t50 nH m0 ee ST Hara cheese everage0t types 40; 207208 Goa 121 B21 Teectoar, dary varia wm “36 @® 1 10 13 NO M0 18 21 32 06 Mi whole evaporated a ee ee + Nak wool pasteurized and steiized ey ee a Parmesan 482 4 1.000 HHD1200 SIMD 19S ISB TS MAD Processed choose plant so oe © ts20 90 600-22 800 2093 102 a2 a7 Ba Sogut, whole ik rut ios “51 20 18 TO 18282887 og whole mk, lan 7% 87 & m0 20 1 17 17 28 32 45 15 ‘Sager proserves, and ewecte (Chocolates mk so 84 = 1mm HCH At AGS Honey 260 Oa Nh 8 8 1B 0208 1-03 Naser caket rc a ee Narmelade 2s On “Ba 06 4 137 00-01 18-18 ‘Suga wre 4900 0 20 2 0 00 Zot ‘egeiabies ‘Asparagus 2% 29 1 mo 7 3 2 w 14 20 98 Brecesl green B44 8% s& 2 Bf 10 22 19 52 Carats. yung 07 © 2 oo 6 2% '% 03 03 56 Cauttower & 38 9 mo of 7 of mB 18 18 74 Coory 7 08 © m ‘5 % 1 02 o1 86 Sheary nu 08 4 i 8 6 2 '% 92 o6 29 Gucumber io 07 2 mo i 8 & 7 03 18 25 Eogpiant is 08 2 mo 10 1 18 i O04 G2 40 (oes, 2 16 2 ko mM |3 wm 9 08 12 3B Lettuce, average ovaries te ooB 3 2 % 6 2% 7 0 05 34 Letuce, aber is oF 2 io i $8 % % 03 03 22 Mushrooms, commen 318 5 0 '6 9 o 6 09 26 49 Onions % 12 8 i 2 4 3 2 06 O7 27 Poppers Capiscum, green 3308 4 WO 8 1 % % o€ o3 21 Palaces og % 24 7 8 6 oF 8 1D 8S Radish rea 07 uM 8's 0 7 sae aa Spinach % 28 0 600 oe ST W118 “prato ioe @ 08 2 mo ‘10 10 19 400 04 03 35 ‘pratoes 7 a7 2 m7 Fm 85 O38 4D 2ucchin 3 18 a eC Ror Nem waar aku Mo = maghesiom P= phosphors, C= chore "ine characterae postabsorpion deteminants et PRAL are aleo preented these are the primasly protén-dopendent urinary sulate excretion: SO,’ the phosphate excess FEX(PEX [meg] PO, Ca Mg): nde alkall excess. ALEX (ALEX miEq]~ Na+ Cl). Each's estimated fam the catesporcing nent {ata byte conversion actors described In Tabi 1; PRAL (mEq oI Cl+PO, +S0,~Na-K--Ca~ Wg) also comesonds 0 SO, + PEK ALEX. ‘Data were dervod tom relerence 02 "Data wore derived rom te marufacture® teature(Apolinars, Bad Neenah Arrweller, Germany Voli, Puy-de Dome, France) ‘Forthoseprocossed (sale) foods for which he tabulated contort dovaied by moce than = 10% from he values determined under the ssaumption of an ‘cquimar Raand Ci content, Cl was calouted Vor th ited Na Gala on an equi bass, Cl(mg)=Na mg) x1 54 ‘Fer eo9 white poten, known o havea parieuaty high methionine and cystone covet @1 5-013 higher conversion factor (i, 0.7382 x 10) was uses to ‘slate renal autate excreton Methionine and cysteine conan lated fo 100 protons approximately 15-0 higher egg white thar og, fr bea (52). 794 / JULY 1995 VOLUME 95 NUMBER 7and renal NAE was assessed using Pearson correlation coefft- cients, and the corresponding regression equation was ob- tained from simple linear regression analysis. The average pif values obtained for definite NAE ranges are given as arithmetic means (Figure 2), The statistical procedures and the caleulations of phosphorus excess, alkali excess, and PRAL. (each per 100 g food) from the nutrient values of the food tables were conducted with the Statistical Package for the Social Sciences (SPSS/PC+, version 4.0, SPSS, Chicago, I). RESULTS ‘The calculated acid-forming potential (or base-forming poten- tial) of more than 100 frequently consumed foods and bever- ages is listed along with postabsorptive urinary determinants, sulfate, phosphorus excess, and alkali excess, according to nine main food groups in ‘Table 2. ‘The calculation model yielded PRAL values ranging fromamaximum of 34.2 mBq/100 8 (parmesan cheese) over 0 mBq/100 g for fat and oils to a minimum of -21 mBq/100 g (raisins). Among the raw Ce, nondried) fruits, the base-forming potential was similar to that, of vegetables. This can also be seen from Table 3, which gives the average PRAL values for certain groups and Subgroups of ods. Along with fruits and fruit juices and vegetables, alkali- rich, low-phosphorus beverages have the lowest (ie, negative) PRAL values, According to Table3, these foods are followed, in the order of gradually increasing average PRAL values, by alkali-poor (low-phosphorus) beverages, fats and oils, milk and, noncheese dairy products, bread, noodles, and flour. Fish, ‘meat-and meat products, and cheeses are the food groupe with, the highest PRAL. ‘Table 4,an example ofan extremely simplified diet, indicates that the exchange of only a few foods (protein-rich or alkali- poor vs alkali-rich) can markedly alter the daily intake of acid equivalents. Table 4 also demonstrates that the renal NAE is, not affected solely by food-dependent acid loads but also by an {important individual factor: the daily excretion rates of organic acids. Use of the average PRAL values (listed in Table 3) instead ofthe data calculated forsingle foodsyielded moderately different estimates for the diet-induced daily acid loads. ‘The correlation found for urine pH and renalNABiis depicted, in Figure 2. The corresponding regression equation and regres- sion ine, as well as the single pH values (each representing the mean of a certain NAE interval of 40 mBq), indicate that consurnption of diets with an estimated renal NAB of about 100, mEq (diet A in Table 4) and about 30 mEq (diet B in Table 4) should result in average 24-hour urine pH of about 5.9 and 6.6, respectively. In other words, corresponding urine pH values are attainable with diets yielding daily PRAL values of either approximately 60 or ~10 mEq/day. DISCUSSION According to the calculation model described, negative PRAL, values (indicating an excess of the base-forming potential of foods) were nearly exclusively found in the vegetable and fruit ‘groups. In contrast, the highest acid loads originated in cheese, followed by meat, fish, and grain products. Similar trends, although with marked deviations for individual foods, were observed by other investigators whose calculations were based. on acid-alkaline ash analyses (13,14) or on current food tables (2). However, none of these researchers considered the (aver- age) net absorption rates of the relevant minerals. In the present study, lower absolute values were found for oth the acid-forming and the base-forming potential, espe- cially for food groups with the highest and the lowest (ie, the negative) acid excess per 100 g food. Thus, compared with the earlier calculations, the total range ofthe potential acid or base FIG 2, Association between urine pH and renal net acid excretion (NAE) in healthy males (adolescents and adults; 2=60) consuming various normal mixed diets. The values above the triangles represent the ‘urine pH means for the respective NAB intervals (each covering 40 mBq) indicated by arrows. Tes Me ton ronal acd loadsPRAL}*o con ood oun and contre tots (lated ts 100g sete porter) OT Food group PRAL (ma) Beverages ‘Akale ih ariow phosphorus? “7 ‘kapoor ara iow phosphorus ° Fats and ole ° Fa 7 Fras and iru juloes™ Ta Grain produce Breas 35 Fou 70 ood, spaghet 67 ost and mest products 38 ‘tic and dary products Mikana ronchoese products® 10 (Chooses wah ower poten contort 20 CChosses wih higher protein coon 238 28 ‘eta reprosent the arthmeto mean of he PRAL vals fhe roopecive foods Sstegin bio? Beverages (phosphorus <30 mq/100_9) wih several tines higher som -rpotasslum content compared ta chloe or example, fe wine, white wna, genan mineral coda) wars and cfs, “Beverages (phosphorus <20 mg/100 g) wih similar soclum + potassium vs hire content. Cocoa (alkali and posphocus-ich) ago fas n this PRAL category Because oe medium phosohorus conten (eg, 28 mg/100g} some Exropean palo bee's havea relatvely nigh PPAL vale (about mEQ/100 9), Swot dred rut ‘respecte ofthe type of four whe meal o wie lan) ‘Primary whey 80088, Suess 189 protein per 100g. 'Nere than 18 9 poten pe 1003 ‘Nee teat ty ow aha exces) ae wprach (ry Noh a JOURNAL OP THE AMERICAN DIETETIC ASSOCIATION / 795Tabled Estimation of daly renal net acid excretion (NAE) for a woman* consuming afcttious diet with a relatively high (diet A) or iow (diet B) potentia! atin of day ene (Nae) onsuming iy igh (diet A) or low (diet B) pote Food Diet A Food Diet intake ‘Proton PRAL®——PRAL® PRAL® intake Energy Protea PRAL RAL PRAL® (o's) (reals) (6/8) (mEq/1009) (megs) (mEq) (1) (eal (G14) (mEGi100g) (MEQ) (megs) Bead ‘wheat Tow 2004682488 76 10 200 466124 Ccatage coe 3504383 35 280 so stat wuiwy 20002489 wee 190 20 214438 Cucumber 200 “20 ta 08. “16 “56 500 “0 Spagheti 120 410A. 7380 a) Bator ‘margarine 102 753 _04 00 00 margarine 147 1085, 2208 i207 44 2206 Daly urinary croton oforpanie aca 2 Daily NAE(estimatec) an agutieraie 63g weight and 163 em neigh SPRAL value akon Wom Table2 ‘SGaledated irom the average PRAL elves listed in Table 9 Estimation of dail excretion of organic acids (OA): (8 (mnie) = Booy surtace area (ne) x41 (MEG. 73 m2) 173 (") “Simple estiation of day xc: of OA using inividual body weight (BW) (} OA (Eich) BW x0 66, Nee ce ee en cc ec ec ee ce a excess of foods is reduced by our physiologically based caleu- lation model. Differences are also discernible for foods from food groups with lower (absolute) PRAL values, for example, rik, -5 0 mEq (14),~2.7 mEq (18), -83 mEq (2), +0.7 mEq (actial value); peas, 1.8 mBq (14), -1.2 mEq (13), +1.2 mEq (actual value); and whole-wheat bread, +7.3 (14), +6.7 mEq (13), +1.8 mBq (actual value). These discrepancies probably aceuirred because none of the former evaluation models took into account the bioavailablity of nutrients Inadaition, the earlier acid-ash diet calculation methodolo- gies did not offer the possibility of predicting the probable Urine pH levels produced by certain diets. As we have shown in healthy men (5), urine pil can be predicted reliably and consequently can be adjusted to a target pH when the PRAL values presented here (for the prognosis of diet-induced NAB) are used with the regression equation for renal NAB and urine pH depictedin Figure2. Although thisregression equation was #blaitted from measurements in men, the corresponding pre- diction of urine pH for women will be valid because a charac- teristic effect of gender on renal NAE does not exist (4,9) IMPLICATIONS AND APPLICATIONS ‘Aformula for the prediction of urine pH using nutrient intake data has been developed for cats (15), a species with a high incidence of struvite stones. To our knowledge no such prog- nostic method has been available hitherto for human beings ‘consuming normal mixed diets. Based on PRAL values pre- sented herein, which consider mean intestinal absorption rates {or individual nutrients and postabsorptive metabolism of sulfur-containing amino acids, it is possible to estimate the diet-dependent component of daily renal NAE, thats, the daily PRAL, This can be achieved by simply adding all single PRAL values of foods and beverages according to their daily ingested amounts (see Table 4). Another NAE component, the excre- tion of organic acid which depends primarily on body weight (or body surface area) and is relatively constant for each Individual, ruust also be considered (Table 4) to yield the total INAE that finally determines each person's urine pH level, 796 / JULY 1995 VOLUME 95 NUMBER 7 ‘No optimal estimation of actual NAB willbe obtained (even based on exact mutrient intake data for the day of urine collection) if dietary composition diffors markedly between the day of the 24-hour urine collection and the preceding day. Our calculation model requires acertain steady state of nutri- ent intake, thats, a relatively constant food supply at least for 2 days. Marked deviations between predicted and measured Urine pH values can also occur for certain persons, or example, as a result of an inherent reduced renal ability to produce ammonium — renal tubular acidosis type IV (16), or depend- ing on the classification, type Il (17). Such a lowered NAE capacity will result, for each acid load (ie, for cach given NAE range), in a markedly lower urine pH level than that depicted inFigure2. This phenomenon which, as a pathological finding, is rae in the nornal population, may occur to a much weaker degree in healthy subjects in addition to the aforementioned sources of inaccuracies ‘and variations involved in predicting urine pli, there may be differences between the tabulated nutrient data (used for PRAL calculation) and the actwal values due to inherent nutri cnt variations ofnatural foods nd differences in their process- ing and preparation (18). Furthermore, the results of predict- ing renal NAE (and consecutively the urine pH of subjects consuming certain diets) depend on accurate food consurp- tion data. In regard to estimation errors related to the normal rutrient variations of foods, it seems appropriate (atleast for ‘rough and rapid survey of'the effects of diet on urine pH) to ‘use the mean PRAL values, which have been averaged for definite foods and food groups in Table 3 Several diseases encountered in dietetics could benefit from application of dietary moans to modify urine pH in people. One of these is urolithiasis, especially in a case of confirmed diag- nosis of caleium phosphate or struvite stones or a case of uric ‘acid lithiasis or eystine stones. In these two cases, a generally ‘accepted basic principle of therapy or recurrence prevention ‘consists of urine acidification or alkalinization, respectively. For magnesium ammonium phosphate (struvite) or caleium phosphate stones, both of which are poorly soluble at higherurine pH values, solubility can be increased and precipitation inhibited by adequate decreases of urine pH. However, no dietary methods to prevent recurrence by improving the solu- bility of these phosphate stones (infective calculi) is possible ‘when patientshave renaltubularacidosis (19). Inpatients with uric acid lithiasis or cystine stones, adjusting urine pH to near 6.8isrecommended. This urine pH level isattainableinhcalthy subjects by purely dietetic measures (4,5). In a case of noninfectious calcium oxalate stones, a pH increase can be of benefit (3,8).’Therefore, the recommended high fluid intake (20) should beachieved by drinkingalkalining, ie, potassium/alkali-rich (low-phosphorus) beverages but not caffee, because coffee contains relatively high amounts of oxalate and excess intake seems to increase renal calcium, losses (21). ‘About 0.5% of all boys and 3% to 6% of all girls (data from, Germany) becomeill from urinary tract infectionbefore reach- ing puberty (22). Increasing resistance to various antibiotics makes dietary treatment of this disorder an option that should be considered. Zimmermann (23) proposed an alteration be- tween several days consuming an acidifying diet and several days consuming an alkali-rich diet; whether this approach is particularly effective in suppressing bacterial growth, and thus in overcoming infection, deserves detailed reinvestigation. Current PRAL data could be valuable in applying the meta- bolic benefits of increased alkali ingestion to athletic perfor- ‘mance in certain sport disciplines (24) and to mineral balance and skeletal metabolism in postmenopausal women (25). Ml Presented, in part, at the 30th scientific congress of the Deutsche Gesellschaft fir Ernahrung, Bonn, Germany, March 1998. This study was supported by the Ménisterium fiir Wissenschaft und Forschung des Landes Nordrhein- Westfalen and by the Bundesministerium far Gesundheit We gratefully acknowledge Mechtild Hohler for excellently organizing and coordinating the ‘investigations in bodybuilders. We also thank Barbara Otto for typing and arranging the tables and Chrisia Chahda for cooperating in the selection of relevant foods and beverages and for performing and checking the correct transfer of the nutrient data from the food tables 10 appropriate data files for the calculation of the potential renal acid loads. References 1, Peacock M, Robertson WG, Pathogenesisofurolithiasi. In-Schneider H.J,ed. Urollttasis: Etiology, Diagnosis. Berlin, Germany’: Springer; 1985:245.39, 2. DuyerJ, FoulkesE, EvansM, Ausman L.Acid/alkaline ash diets: time {or assessment and change. J Am Diot Assoc. 1985; 85:841-845, 8. Sioncr R, Hesse A. BinfluS verschledencr Kostformen auf dle Hormausammensetzuny und des Kaliumoxalat Steinbildungsrsiko, Z Bruahrangswiss. 1988; 3246-65, 4. Romer’, Mang . Estimation of therenalnctacid exeretion by aduls ‘consuming diets containing variable amounts of protein. Am «I Clin ‘Nutr. 1994; 59:1366-1361, 5. RemerT, ManzF. Dietary protein as amodulator ofthe renal netacid capacity: evidence that an increased protein Intake improved the capability of the kidney to exerete ammonium. J Nutr Biochem. in Prose 6. Manz ¥, Schmidt H, Scharer K, Bickel H. Acid-base status in dictary seatment of phenyiKetonuri. Pediatr Res. 1977, 11:1084-1087, 7. Kalhoff H, Manz ¥, Diekmann L, Stock GJ. Suboptimal mineral composition of cow's milk formulas: isk factor for the development of Tate metabolic acidosis. Acta Paediatr Scand. 1990; 79:748-T49. 8. Manz P, Schunidt HL Retrospective approach to explain growth retardation and urolithasis ina child with long-term nutritional acid loading. Z Brnahrungswiss. 1992; 31:121-129. ‘Manz F, Veesei P, Wosch H. Renale Stureansscheidung und renale Molonlast bel gesunden Kindern und Erwacksenen. Monatsschr Kindorheilkd. 1984; 122:168-167. 410. Holland B, Weieh AA, Unwin ID, Buss DH, Paul AA, Southgate DAT. McCance and Widdowson's the’ Composition of Foods. 6th ed. Cambridge, England: The Royal Society of Chemistry; 1991 LL, Hobler M, Decher-Spliethot F, Kersting M, Temes ML, Marz F. Punktionsbelastung es Stoftwechstls und der Niere bei Krafsportiorn rit eiweitreicher Kost. Disch 2 Sportmed. 1984, 4:92-103. 12. Manz, Romer'?, Decher SpliethofT &, Hohler M,Kersting M, Kunz , Lausen H. fects fa high protein intake on renal cid excretion in Doxiybullders. Z Ernahrangswiss. 1995; 3410-15. 413. Gonick HC, Goldberg G, Muleare D. Reexamination f the acidash ‘content of several diets. A J Clin Nut. 1968; 21:898-905, 1A, Langendor? H. Saure-Basen-Gleichgewicht utd chronische acldogene und alkalogene Emnatung, Z Eruakrungswiss. 196 2 (supp) 1-93. 15. KiensleB, Schublenocht A. Intersuchungen zur Struvitsteindiatotik 1. Binflu® vorschiodener Futierrationen auf den Harn-pH-Wert der Katze. DTW. 1983; 100:198-208 16. Schurer K, Mary F. Hereditare Tubulopathion. In Losse H, Renner E, eds. Klinische Nephrotogio. vol 2. Stuttgart, Germany: Thieme; 1982:912-949, 17. Carlisle FSF, Donnelly SM, Halperin ML, Renal tubular acidosis (RETA): Recognize the Amunonium defect and pHonget the urine pi. Padiatr Nephrol. 1991; 5:242-248, 438, Pennington JAT, Wilson DB. Dally intakes of nine nutritional tlements: analyzed vs calculated values. Am Diat Assoc. 1900;00:375- 81 19, Lutzeyer W, Hering F: Drug therapy of uritary calculi and preven: ton of recurrence. in: Schneider HJ, ed. Urolithiasis: Therapy, Prevention, Belin, Germany’ Springer, 1985:1.73. 20. Lemann J. Composition of tne diet and calcium kidney stones. NV Engl J Mod. 1998; 328:80-881 21, Hasling, Sondergaard K, Charles P, Moselle L, Calelum metabo- lism in postmenopausal women i determined by dictary ealelum ad coffee intake, J Nutr 1992; 122-1119-1 12 22, Schofer 0, Beotz R, Mannhardt W, Schulte- Wissermann H. Welehe ‘Ari von Abwelirachwiahe besteht bei Kindern mit rezidviorenden, nicht-obstruktiven amwegsinfektionen? Kinderarst, 1999; 24:715: TA. 23. Zimmermann W. Natiliche Urologiks: Hlarnsteinproplylaxe, Hiarnwegainfekte, Ditormen, Tel 1. Fortschr Med. 1988; 1061820, 24, Gao dP, Costil DL, Horswill CA, Park SH. Sodium bicarbonate ingostion improves performance in interval swirumin. Bur J Appl Physiol 1988; 58:171-174 25, Sebastian A, Harris ST, Ottaway JH, Todd KM, Morris RC. Improved rijeral balance and skeletal 1etabolism in postmenopausal wornen fronted with potassium bicarbonate. N Bgl J Med. 1904, 2061776 181, 26. Documenta Geigy. Wissonschaftiche Tabelien. 7th ed. Stttgart, many: Thieme, 1975:688-461, 27. Kesleloot Hy, Joossens.1V. The relationship between dictary intake ‘and urinary excretion of sodium, potassium, calcium and mayavesiu: Holgian interuniversity rescarch on nutrition and health. J Hum Hypertens. 1980, 4527-533, 28. Stanbury SW. Intestinal absorption of calc and phospliorus in adltmanin heal anu disease. In: Bickel, Stern J, eds. aborn Brors ©f Calcium and Hone Metabolism. London, England: MTP Press, lare2i-46, 29, Maschio G, D'Angelo A, Mioni (, Ossi EIT metabolisme idvo- fletrotitica nelle malate’ del rene. Padova, Naly: Pein Editore; 1972:60-68, 0, Schulian JD, Sulfurmetabols In Schulman JD, ed, Cystinosis. Washington, DC: Govertusent Printing Office; 1973. Dept of Health, Education, and Welfare publication NIH 72-249. BL, Food and Nutrition Board, Recommended Dietary Allowances. 430th ed, Washington, DC: National Acaderty Press; 1989:32:33, 32, Soul SW, Fachmann W, Kraut H. Food Composition and Nutri- tion Tables 1989/90. 4th ed. Statgart, Germany: Wissenschaflice Verlagsgescllschaft; 1989 JOURNAL OF THE AMERICAN DIETETIC ASSOCIATION / 787