The Journal of Nutrition Vol. 127 No. 7 July 1997, pp. 1279-1283
Copyright ©1997 by the American Society for Nutritional Sciences

Low Dietary Protein Impairs Blood Coagulation in BHE/cdb Rats^1,2

Yow-Ling Chang, Hee-Sook Sohn^*, Kung-Chi Chan, Carolyn D. Berdanier, and James L. Hargrove³

The Department of Foods and Nutrition, The University of Georgia, Athens, GA 30602-3622, ^* Department of Food Science and Human Nutrition, Chonbuk National University, Chonju 561-756, Korea and  Department of Foods and Nutrition, Providence University, Shalu, Taichung, Taiwan, ROC

ABSTRACT

INTRODUCTION

MATERIALS AND METHODS

RESULTS

DISCUSSION

ACKNOWLEDGMENTS

FOOTNOTES

LITERATURE CITED

ABSTRACT

The influence of dietary protein on blood coagulation tests was evaluated in BHE/cdb rats. Three experiments were conducted in order to compare effects of diets with low (8 g/100 g diet) or high (38 g/100 g diet) protein, to establish values for coagulation tests at intermediate (12-30 g/100 g diet) concentrations of dietary protein, and to compare feeding identical quantities of diets with 8 g protein/100 g diet vs. 18 g protein/100 g diet. After 4 wk of feeding the semipurified diets, bleeding time exceeded 15 min in the groups fed low protein diets, compared to a range of 3-6 min for the groups fed high protein diets. Several in vitro tests of coagulation were abnormal in the rats fed low protein diets. For example, prothrombin time averaged 27 ± 8 s in rats fed 8 g protein/100 g diet plus beef tallow, but 17 ± 1 s in rats fed 38 g protein/100 g diet plus tallow. The coagulation deficit in rats fed low protein was not affected by fat source (tallow vs. menhaden oil), but fibrinogen was elevated in rats fed diets with menhaden oil. Conversely, no differences in coagulation tests were observed among rats fed 12-30 g protein/100 g diet. Bleeding times ranged from 7 to 9 min, and prothrombin time was 17-18 s. Significant differences in plasma fibrinogen concentration and prothrombin time were observed in rats fed 8 vs. 18 g protein/100 g diet at a fixed rate of 6 g/100 g body weight. Platelet and blood cell numbers were unaffected by dietary protein. The evidence for multiple deficits in the coagulation system suggests that hepatic function in BHE/cdb rats may become impaired when the rats are fed low protein diets of the composition used here.

INTRODUCTION

Blood coagulation is controlled by a balance between procoagulant and anticoagulant systems in the blood plasma, the platelets, the vascular wall, and tissue factors that are exposed by injury (Eastham and Slade 1992, Halkier 1991). Several nutrients affect coagulation, including vitamin K, which is needed by the liver for synthesis of prothrombin and other coagulation factors that are carboxylated post-translationally (Suttie 1991). In addition, the substitution of long chain, (n-3) polyunsaturated fatty acids for arachidonic acid in platelet membrane phospholipids can contribute to an abnormally prolonged bleeding time due to altered platelet function (Beswick et al. 1991, Chin 1994, Dyerberg et al. 1978). An increase in total dietary fat may affect coagulation by activating factor VII zymogen in the blood plasma (Miller et al. 1986, Miller 1992).

The effect of protein nutrition on coagulation has not been studied extensively. However, during a study of interactions between dietary protein and fat type on kidney function, Chan (1993) observed qualitatively that blood coagulated more rapidly in BHE/cdb rats fed 38 g protein/100 g diet compared to 8 protein/100 g diet. The BHE/cdb rat is a non-obese animal model of noninsulin-dependent diabetes mellitus (NIDDM)⁴ that develops diabetes as it ages (Berdanier 1991 and 1995) and has a defect in the mitochondrial ATPase gene (Mathews et al. 1995). Due to the design of the prior study (Chan 1993), it was not clear whether the high protein diet accelerated coagulation, or whether the low protein diet impaired coagulation. Thus, this study was conducted to compare effects on several laboratory tests of coagulation of feeding BHE/cdb rats low, moderate and high levels of dietary protein.

---

MATERIALS AND METHODS

Animal care and diet composition. Specific pathogen-free, male BHE/cdb rats from the colony at the University of Georgia (Berdanier 1995) were used according to procedures for care and experimental protocols approved by the UGA Institutional Animal Care and Use Committee. The rats were 2-3 months old at the beginning of the experiments, and were housed individually in wire-bottom cages in a room controlled for temperature (21 ± 1°C), humidity (40-50%), and light (lights on, 0600-1800 h). Diets were isocaloric, with sucrose substituted to compensate for differences in protein (casein:lactalbumin, 1:1 by weight), and fat was contributed by 1 g corn oil/100 g diet and either 9 g beef tallow or 9 g menhaden oil/100 g diet, as indicated (Table 1).

Table 1. Diet composition1 [View Table]

Experiment 1. The first experiment was designed to identify interactions between protein concentration and fat type on coagulation. For one week, 32 rats were fed a semipurified diet that contained (per 100 g diet) 8 g casein-lactalbumin, 9 g beef tallow and 1 g corn oil, and food intakes were recorded. The rats were then randomly assigned to four groups (8 rats each) and fed diets containing 8 or 38 g protein/100 g diet, with fat provided by 1 g corn oil plus either 9 g beef tallow or 9 g menhaden oil/100 g diet. Different fat sources were used solely to duplicate the diets in the preliminary study (Chan 1993). Body weights and food and water intakes were measured weekly.

Experiment 2. The second experiment was conducted to establish norms for coagulation as a function of dietary protein concentration. One group of 7 rats was fed a nonpurified diet (Purina Rodent Chow, Richmond, IN), and four groups of 7 rats each were fed semipurified diets containing 12-30 g protein/100 g diet, with fat provided by 1 g corn oil and 9 g beef tallow/100 g diet.

Experiment 3. The last experiment was designed to compare effects on coagulation tests of feeding rats diets with either 8 or 18 g protein/100 g diet. However, rats in the low protein group initially consumed about 1 g more food per 100 g body weight per day than did the rats fed 18 g protein/100 g diet, when food was freely available. To equalize food intake, both groups were fed at the rate of 6 g diet/100 g body weight per day after the third week, when tests of bleeding time were carried out as described (Tschopp and Zucker 1972). Because the bleeding test could potentially activate the coagulation system, the rats were allowed to recover for the final two weeks. After a total of 5 wk, the rats were anesthetized with ketamine:xylazine (10:1), and blood samples for coagulation tests were collected from the inferior vena cava as described previously (Chan et al. 1993).

Blood collection and coagulation tests. Blood was collected from the inferior vena cava of anesthetized rats into syringes containing 0.5 mL of disodium citrate (0.16 mol/L) at a ratio of 9 volumes of blood:1 volume disodium citrate. To prepare platelet-rich plasma, whole blood was centrifuged at 190 × g for 15 min at 22°C, and the supernatant was collected. The recalcification test was carried out by adding one-tenth volume of 25 mmol calcium chloride/L to fresh platelet-rich plasma, and measuring the time until the initial clot was formed with a fibrometer (COAG-A-MATE^® XM, Organon Teknika, Durham, NC). The other coagulation tests used platelet-poor plasma, which was prepared by centrifugation of platelet-rich plasma at 2000 × g for 20 min at 4°C. Platelet-poor plasma was either used fresh or frozen in aliquots at 70°C. For later tests, each portion was thawed once and then discarded. The protocol for recalcification of platelet-poor plasma was the same as for platelet-rich plasma. The assays for prothrombin time (PT), activated partial thromboplastin time (APTT), and fibrinogen were performed on an automated fibrometer COAG-MATE^® XC-2000 (Organon Teknika) in the College of Veterinary Medicine at the University of Georgia. PT was determined by mixing plasma with a thromboplastin reagent (Simplastin^® Excel, Organon Teknika) and timing formation of the initial clot. APTT was determined by mixing sample plasma with 25 mmol calcium chloride/L and a partial thromboplastin reagent (Automated APTT, Organon Teknika) and timing initial clot formation. Similarly, the fibrinogen concentration was determined by mixing diluted sample plasma with a thrombin reagent (Fibriquik^® thrombin reagent, Organon Teknika) and timing formation of the initial clot. The factor VII assay was carried out on the COAG-A-MATE^® XM fibrometer by mixing the diluted plasma with factor VII deficient plasma (Organon Teknika Co.), and timing clot formation after thromboplastin reagent (Thromboplastin with calcium, Sigma, St. Louis, MO) was added. Coagulation controls were performed using Level I normal human plasma (Sigma) for the PT, APTT, and fibrinogen tests. The tests of prothrombin time and activated partial thromboplastin time for Experiments 1 and 2 were done at the same time using platelet-poor plasma that had been stored at 70°C. The tests were conducted by one operator using the same set of reagents.

Plasma total protein concentration was determined with a kit based on the bicinchoninic acid reagent with bovine serum albumin as a protein standard (Pierce Chemical, Rockford, IL). Albumin concentration was analyzed with a bromcresol green kit (Sigma Diagnostics kit 631-2). Glucose was assayed with kit 115-A based on glucose oxidase, and urea was quantified using kit 640-A based on urease (Sigma Diagnostics).

Statistics. Data were analyzed with the Tukey-Kramer multiple comparisons after 2-way ANOVA using the Statistical Analysis System computer program (SAS 1989, SAS/STAT Version 6, SAS Institute, Cary, NC) in Experiment 1, and Dunnett's multiple comparisons test in Experiment 2 (Steel & Torrie, 1960). Because the variances for PT and APTT were not equal in Experiment 1, means were compared with the Wilcoxon signed rank test (Steel & Torrie, 1960). Means were compared using Student's t test in Experiment 3. Variation in data is expressed as means ± SD (n = 7-10). Differences were considered significant at P  0.05.

---

RESULTS

Blood coagulation in relation to dietary protein and fat type. Rats fed the low protein diets consumed more food than did the high protein groups [6.26 ± 0.68 vs. 5.28 ± 1.37 g/(100 g body weight·d), respectively], but weighed significantly less at autopsy (Table 2). Bleeding time was significantly longer in the low protein diet groups compared to the high protein diet groups regardless of fat source. In vitro measures of coagulation that differed in the low protein diet groups included clotting time (as tested by recalcification of platelet-rich plasma), PT and APTT (Table 2). Plasma albumin concentrations did not differ among groups (not shown). Significantly higher fibrinogen concentrations were found in the groups fed menhaden oil compared to those fed corn oil. The activity of factor VII was significantly lower in low protein diet groups than in high protein diet groups irrespective of the fat source, as indicated by the longer periods required to generate fibrin in the low protein groups (Table 2). The statistical tests suggested that fat interacted significantly with protein in modulating PT. However, the variances for PT and APTT in the low protein and high protein groups were not equal. The data were therefore analyzed using the Wilcoxon signed rank test. By this measure, significant differences were observed for PT and APTT for groups fed 8 versus 38 g protein/100 g diet with beef tallow, or 8 versus 38 g protein/100 g diet with menhaden oil (Table 2, footnote). Comparisons between groups fed diets with tallow or menhaden oil of the same protein content were not significant.

Table 2. Effects of two dietary protein levels and two fat sources on body weight and the blood coagulation system in BHE/cdb rats1 [View Table]

Blood coagulation in relation to dietary protein content. In Experiment 2, growth rates and the total change in body weight did not differ among groups, nor did intakes of food or water differ significantly (data not shown). To facilitate comparisons with Experiment 1, the tests for PT, APTT, and fibrinogen shown in Tables 2 and 3 were done at the same time using samples that had been stored at 70°C. In Experiment 2, the bleeding time, clotting time, PT, APTT, and fibrinogen concentration did not differ significantly among groups with dietary protein concentrations ranging from 12 to 30 g/100 g diet (Table 3).

Table 3. Effects of dietary protein on the blood coagulation system in BHE/cdb rats [View Table]

Blood coagulation in rats fed equal amounts of diet. The results of the first two experiments suggested that coagulation was normal when rats were given free access to diets with protein concentrations of 12-38 g/100 g diet, but was abnormally prolonged when protein was decreased to 8 g. A third experiment compared the effects of feeding fixed amounts of diets with 8 and 18 g protein/100 g diet with 9 g beef tallow and 1 g corn oil/100 g diet. Rats fed 8 g protein/100 g diet initially consumed significantly more food each day than did the group fed 18 g protein/100 g diet (7.61 ± 0.70 vs. 6.23 ± 0.84 g/100 g body weight, respectively; P < 0.005). Therefore, during the final two weeks, rats in both groups were fed 6 g of diet per 100 g body weight. Fibrinogen concentration was significantly lower in rats fed low protein diet compared to high protein diet (1.62 ± 0.36 vs. 2.26 ± 0.35 g/L, respectively; P = 0.005), and prothrombin time was also significantly prolonged in the low protein group (19 ±2 vs. 16 ± 1 s, respectively, P < 0.005). The coagulation test with recalcified, platelet-rich plasma was slightly longer in samples from the low protein group (95 ± 15 vs. 80 ± 15 s, P = 0.05), but the values for platelet-poor plasma did not differ. In addition, APTT, plasma total protein and albumin concentrations, and numbers of red blood cells, white blood cells, and platelets did not differ between groups (not shown). Plasma glucose concentration was higher in the low protein group (7.3 ± 1.5 vs. 6.5 ± 0.6 mmol/L), but the difference was not significant. Plasma urea concentration was significantly lower in the rats fed low protein diet than in those fed high protein diet (1.32 ± 0.40 vs. 2.44 ± 0.52 mmol/L, respectively, P < 0.005).

---

DISCUSSION

This study supports the hypothesis that the coagulation system in the diabetes-prone BHE/cdb rat becomes impaired when the rats are fed 8 g protein/100 g diet. Conversely, the data fail to demonstrate any effect of dietary protein within the range of 12 to 38 g casein-lactalbumin/100 g diet. Experiment 2 provided normal values for several coagulation tests in BHE/cdb rats. Compared to these values, only the two groups fed 8 g protein/100 g diet in Experiment 1 differed greatly; the two groups fed high protein diet in Experiment 1 had normal bleeding times and prothrombin times. Impaired coagulation was observed after approximately 4-5 wk of feeding. Even within the groups fed 8 g protein/100 g diet, not all animals showed abnormal coagulation, which gave rise to large sample variance. This suggests that some animals may have compensated well to the low protein diet, whereas others failed to compensate adequately. It is unclear why a less extreme coagulation deficit was observed when rats were fed a fixed amount of diet for the last two weeks, instead of allowing free feeding (Experiment 3). However, if the impaired coagulation is due to low protein intake, the deficit should become more severe at concentrations of protein below 8 g/100 g diet. This would be consistent with the observation that coagulation was normal at all higher concentrations of dietary protein.

The prolonged bleeding time paralleled several in vitro coagulation tests, including: 1) abnormal coagulation after recalcification of either platelet-rich or platelet-poor blood plasma; 2) decreased fibrinogen concentration; 3) prolonged prothrombin time and activated partial thromboplastin time; and 4) a significant decrease in factor VII activity. The results of the in vitro tests of coagulation suggest that the primary defect in coagulation persisted in platelet-poor plasma, and was attributable to altered activity of the soluble coagulation system. Affected factors included fibrinogen, prothrombin, thromboplastin and factor VII. Thus, the low protein diet produced a general coagulopathy, rather than specifically altering one component of this system. Although fish oils reduce the rate of platelet aggregation and prolong bleeding time in humans (Chi 1994) and rats (Song and Wander 1991), the low protein group had prolonged bleeding time even when fed tallow as a fat source. It was surprising that no significant difference in bleeding time was noted between rats fed diets with beef tallow as compared to menhaden oil, but the tail bleeding test is not a sensitive indicator of platelet function.

The content of high-quality protein that will maintain health in a normal, mature rat is approximately 6 g/100 g diet when other constituents are present at recommended levels (NRC 1978). The protein concentration of 8 g/100 g diet used in this study was substantially higher than the 0-5 g/100 g diet protein concentrations used in attempts to produce experimental protein deficiency in rodents (Anthony and Edozien 1975, Lago et al. 1993). At protein concentrations below 5 g/100 g diet, food intake is reduced and no growth is observed, in contrast to the increased food intake and moderate growth noted here in rats fed protein at 8 g/100 g diet. In addition, plasma albumin and plasma total protein were normal in these rats, and while weight gain was modest, there were no signs of frank protein deficiency. Thus it appears that BHE/cdb rats develop hepatic dysfunction as indicated by reduced concentrations of plasma coagulation factors when fed diets containing 8 g protein/100 g diet, with other constituents as described.

The BHE/cdb strain is a non-obese model of NIDDM that develops abnormal glucose tolerance and lipemia at approximately 300 d of age (Berdanier 1995). This disorder can be elicited at younger ages by feeding diets with high sucrose concentrations (65 g/100 g), which increase synthesis of fatty acids and liver triglyceride (Berdanier 1991). The livers of BHE/cdb rats develop fatty infiltration under the influence of high carbohydrate diets (Marshall et al. 1969), consistent with altered mitochondrial metabolism (Berdanier 1995). Noninsulin-dependent diabetes mellitus alters liver function and blood coagulation in human patients (Donders et al. 1993), and the data of Experiment 3 suggest that the rats used here were beginning to develop poor glucose tolerance (plasma glucose concentration, 7.3 ± 1.5 vs. 6.5 ± 0.6 mmol/L in the 8 and 18 g protein/100 g diet groups, respectively; P = 0.11). It should be noted that a coagulation deficit is an important general sign of liver dysfunction (Munoz et al. 1988). Consistent with this idea, a subsequent study showed that BHE/cdb rats fed diets containing 6 g protein/100 g diet supplemented with 0.25 g choline bitartrate/100 g diet develop midzonal necrosis (J. L. Hargrove, unpublished observation).

The finding that the low protein diet produced a coagulation deficit is similar to reports that the soluble coagulation system is usually normal in human subjects with marasmic protein-energy malnutrition, but abnormal in kwashiorkor (Hassanein and Tankovsky 1973, Kalala et al. 1983). Bleeding disorders are not a constant feature of protein-energy malnutrition (Keys et al. 1950, Torun and Chew 1994, Viteri et al. 1968), and would not be expected unless the liver or platelets were affected. Two signs of impaired liver function in human patients with protein-energy malnutrition are fatty infiltration and edema (Flores et al. 1974, Truswell et al. 1969). These symptoms suggested to Jackson (1990) that kwashiorkor represents a failure of the liver to adapt to malnutrition and other environmental stressors, whereas marasmus represents a condition of better adaptation. The data presented here show that an imbalanced diet with low protein and high carbohydrate also impairs liver function in the BHE/cdb rat in the absence of environmental stressors such as enteritis. Because the concentration of albumin and other plasma proteins declines in protein deficiency (Waterlow 1992, Young and Marchini 1990), it is reasonable to expect that concentrations of coagulation factors may be depressed as well. Similarly, Meghelli-Bouchenak et al. (1987) have shown that low protein diets alter lipoprotein secretion, a feature that coincides with or precedes steatosis. Nutritional liver injury is produced when rats are fed diets that contain elevated sucrose (Bacon et al. 1984), imbalanced amino acid profiles (Newberne et al. 1969), or insufficient methyl donors (Zaki et al. 1963). Given this background, it seems likely that the coagulation deficits noted here represent one symptom of a general loss of liver function, rather than a specific effect of protein deficiency. If this conclusion is correct, other markers of hepatic injury should be affected in addition to coagulation. We are now testing the hypothesis that low protein intake promotes liver dysfunction in metabolically normal rats without necessarily producing steatosis.

---

ACKNOWLEDGMENTS

We thank Diane Hartle and Michelle Barton for procedural advice and use of equipment and facilities during this study.

FOOTNOTES

Manuscript received 8 August 1996. Initial reviews completed 9 October, 1996. Revision accepted 4 March, 1997.

LITERATURE CITED

• American Institute of Nutrition Report of the American Institute of Nutrition ad hoc committee on standards of nutritional studies. J Nutr. 1977; 107:1340-1348
• Anthony L. E., Edozien J. C. Experimental protein and energy deficiencies in the rat. J. Nutr. 1975; 105:631-648
• Bacon B. R., Park C. H., Fowell E. M., McLaren C. E. Hepatic steatosis in rats fed diets with varying concentrations of sucrose. Fund. Appl. Toxicol. 1984; 4:819-826 [Medline]
• Berdanier, C. D. (1995) NIDDM in the non-obese BHE/cdb rat. In: Lessons from Animal Diabetes (Shafrir, E., ed.) pp. 231-246. Smith Gordon, Ltd., London, England.
• Berdanier C. D. The BHE rat: an animal model for the study of noninsulin-dependent diabetes mellitus. FASEB J. 1991; 5:2139-2144 [Abstract]
• Beswick A. D., Fehily A. M., Sharp D. S., Renaud S., Giddings J. Long term diet modification and platelet activity. J. Int. Med. 1991; 229:511-515[Medline]
• Chan, K. C. (1993) Effects of dietary protein on renal hypertrophy and coagulation in rats. Doctoral dissertation, The University of Georgia, Athens, Georgia.
• Chan K. C., Lou P. P., Hargrove J. L. High casein-lactalbumin diet accelerates coagulation in rats. J Nutr. 1993; 123:1010-1016
• Chin, J.P.F. (1994) Marine oils and cardiovascular reactivity. Prostaglandins, Leukotrienes, and Essential Fatty Acids 50: 211-222.
• Donders S. H., Lustermans F. A., van Wersch J. W. Glycometabolic control, lipids, and coagulation parameters in patients with noninsulin-dependent diabetes mellitus. Int. J. Clin. Lab. Res. 1993; 23:155-159 [Medline]
• Dyerberg J., Bang H. O., Stoffersen E., Moncada S., Vane J. R. Eicospentaenoic acid and prevention of thrombosis and atherosclerosis? Lancet 1978; 2:117-119 [Medline]
• Eastham, R. D. & Slade, R. R. (1992) Clinical Haematology. Butterworth-Heinemann, Ltd., Oxford, England.
• Flores H., Seakins A., Brooke O. G., Waterlow J. C. Serum and liver triglycerides in malnourished Jamaican children with fatty liver. Am. J. Clin. Nutr. 1974; 27:610-614 [Abstract]
• Halkier, T. (1991) Mechanisms in coagulation, fibrinolysis, and the complement system, Cambridge University Press, Cambridge, England.
• Hassanein E. A., Tankovsky I. Disturbances of coagulation mechanism in protein-calorie malnutrition. Trop. Geograph. Med. 1973; 25:158-162
• Jackson, A. A. (1990) The aetiology of kwashiorkor, In: Diet and Disease in Traditional and Developing Societies (Harrison, G. A. & Waterlow, J. C., eds.), pp. 76-113. Cambridge University Press, Cambridge, England.
• Kalala T., Capel P., Hariga-Muller C., Fondu P. Etude de l'allongement du temps de thrombine observe dans la malnutrition proteo-energetique. Ann. Soc. Belge Med. Trop. 1983; 63:233-240[Medline]
• Keys, A., Brozek, J., Henschel, A., Mickelsen, O. & Taylor, H. L. (1950) The Biology of Human Starvation. University of Minnesota Press, Minneapolis, MN.
• Lago E. S., Teodosio N. R., Araujo C. R., Azevedo M. C., Pessoa D. C., Campos F. A., Zucas S. M., Flores H. Rat models of protein and protein energy malnutrition. Int. J. Vitam. Nutr. Res. 1993; 63:52-56 [Medline]
• Marshall M. W., Smith B. P., Lehman R. P. Dietary response of two genetically different lines of inbred rats: lipids in serum and liver. Proc. Soc. Exp. Biol. Med. 1969; 131:1271-1277 [Medline]
• Mathews, C. E., McGraw, R. A. & Berdanier, C. D. (1995) A point mutation in the mitochondrial DNA of diabetes-prone BHE/cdb rats. FASEB J., 9:1638-1642.
• Meghelli-Bouchenak M., Boquillon M., Belleville J. Time course of changes in rat serum apolipoproteins during the consumption of different low protein diets followed by a balanced diet. J. Nutr. 1987; 117:641-649
• Miller G. J. Environmental influences on hemostasis and thrombosis. Diet and smoking. Ann. Epidemiol. 1986; 2:387-397
• Miller G. J., Martin J. C., Webster J., Wilkes H., Miller N. E., Wilkinson W. H., Meade T. W. Association between dietary fat intake and plasma factor VII coagulant activity  a predictor of cardiovascular mortality. Atherosclerosis 1992; 60:269-277
• Munoz S. J., Maddrey W. C. Major complications of acute and chronic liver disease. Gastroenterol. Clin. North Amer. 1988; 17:265-287[Medline]
• National Research Council. (1978) Nutrient requirements of laboratory rats. pp. 12-18. National Academy of Sciences, Washington, D.C.
• Newberne P. M., Rogers A. E., Bailey C., Young V. R. The induction of liver cirrhosis in rats by purified amino acid diets. Cancer Res. 1969; 29:230-235 [Abstract/Free Full Text]
• Song J., Wander R. C. Effects of dietary selenium and fish oil (MaxEPA) on arachidonic acid metabolism and hemostatic function in rats. J. Nutr. 1991; 121:284-292
• Steel, R.G.D. & Torrie, J. H. (1960) Principles and Procedures of Statistics, pp. 111-114 and 402-403, McGraw-Hill, New York, NY.
• Suttie, J. W. (1991) Vitamin K. In: Handbook of Vitamins, (Machlin, L.J., ed.) 2nd ed., Marcell Dekker, New York, NY.
• Torun, B. & Chew, F. (1994) Protein-energy malnutrition, In: Modern Nutrition in Health and Disease, (Shils, M. E., Olson, J. A. & Shike, M., eds.) vol. 2, Lea & Febiger, Malvern, PA.
• Truswell A. S., Hansen J. D., Watson C. E., Wannenburg P. Relation of serum lipids and lipoproteins to fatty liver in kwashiorkor. Am. J. Clin. Nutr. 1969; 22:568-576 [Abstract]
• Tschopp T. B., Zucker M. B. Hereditary defect in platelet function in rats. Blood 1972; 40:217-226 [Abstract/Free Full Text]
• Viteri F. E., Alvarado J., Luthringer D. G., Wood R. P., II. Hematological changes in protein calorie malnutrition. Vitamins Hormones 1968; 26:573-615[Medline]
• Waterlow, J. C. (1992) Protein Energy Malnutrition, pp. 97-99, Edward Arnold, London, England.
• Young V., Marchini J. S. Mechanisms and nutritional significance of metabolic responses to altered intakes of protein and amino acids, with reference to nutritional adaptation in humans. Am. J. Clin. Nutr. 1990; 51:270-289 [Abstract/Free Full Text]
• Zaki F. G., Bandt C., Hoffbauer F. W. Fatty cirrhosis in the rat. III. Liver lipid and collagen content at various stages. Arch. Pathol. 1963; 75:648-653

0022-3166/97 $3.00 ©1997 American Society for Nutritional Sciences