- Email: [email protected]
Dietary habits and nutritional status of renal transplant patients
ORIGINAL RESEARCH
Dietary Habits and Nutritional Status of Renal Transplant Patients James Heaf, MD,* Ulla Jakobsen,* Erling Tvedegaard, MD,* Inge-Lis Kanstrup, MD,† and Niels Fogh-Andersen, MD‡ Background: Although dialysis nutritional problems are well described, nutritional problems after renal transplantation (RT) have received little attention. Methods: Body composition as assessed by dual-energy x-ray absorptiometry in 115 stable patients 6.6 ⫾ 5.9 years after RT and repeated 2.9 years later, when a 3-day dietary history was obtained in 79 patients. Results: Patients diet was generally sufficient, but was characterized by a high fat intake and deficiencies in folic acid, vitamin D, thiamine, iodine, selenium, and iron intake. Patients were often overweight, and at any given weight had a 4% to 5% higher proportion of body fat than normal. Loss of fat weight was related to high initial fat weight, long RT duration, and low plasma bicarbonate, but not steroid dose. Conclusion: Dietary advice concerning fat intake is indicated for RT patients, and nutritional supplements with folic acid and vitamin D are generally required. Their main nutritional problem is obesity. This is not adequately measured by body mass index, which should be supplemented by dual-energy x-ray absorptiometry. Attention should be paid to the prevention of acidosis. © 2004 by the National Kidney Foundation, Inc.
A
LTHOUGH THE NUTRITIONAL STATUS of dialysis patients is well described, few studies concerning patients after renal transplantation (RT) have been performed,1-4 and these were usually conducted in the immediate postoperative period. We decided to perform a study of stable long-term RT patients to assess dietary habits, nutritional status as judged by body composition and plasma proteins, and possible causal relationships among these.
Patients and Methods Patients A total of 135 patients with a functioning RT were included in a cross-sectional study of body *Department of Nephrology, Copenhagen University Hospital in Herlev, Herlev, Denmark. †Department of Clinical Physiology, Copenhagen University Hospital in Herlev, Herlev, Denmark. ‡Department of Clinical Biochemistry, Copenhagen University Hospital in Herlev, Herlev, Denmark. Address reprint requests to James Heaf, Graevlingestien 9, 2880 Bagsvaerd, Denmark. Email: [email protected] © 2004 by the National Kidney Foundation, Inc. 1051-2276/04/1401-0004$30.00/0 doi:10.1053/j.jrn.2003.09.005
20
composition. Three years later, 120 patients, including 5 patients not in the original study, were restudied. The patients gave informed consent in accordance to the Helsinki 2 declaration. Immunosuppressive therapy was uniform during the period of study. Transplanted recipients were treated with prednisone 1.5 mg/kg after transplantation, tapered to 20 mg daily after 20 days and 5 to 10 mg after 12 months, and with azathioprine 2 mg/kg/d tapered to 1 mg/kg/d after 20 days. Microemulsion formulation cyclosporine A (CyA) was given twice daily to achieve a trough level of 200 to 250 ng/mL (specific assay; Abbott, Abbott Park, IL), tapered to 100 to 150 ng/mL after 6 months. Antilymphocytic immunoglobulin (initially Minnesota ALG, later ATGAM) was included as induction therapy. Tacrolimus (instead of cyclosporine) and mycophenolate mofetil (instead of azathioprine) were given to occasional patients with a complicated postoperative course.
Methods The following clinical data were registered: age, sex, height, renal diagnosis, transplantation duration, and average dose of prednisone. The Journal of Renal Nutrition, Vol 14, No 1 ( January), 2004: pp 20-25
RENAL TRANSPLANTATION AND NUTRITION
following investigations were performed and were repeated 3 years later. 1. Plasma albumin, protein, creatinine, bicarbonate, urea (second analysis only), magnesium (second analysis only). 2. A 24-hour urine collection was analyzed for volume, creatinine, creatinine clearance, and urea (second analysis only). 3. The following variables were measured using dual-energy x-ray absorptiometry (DEXA; DPX-IQ, Lunar Radiation Corporation, WI): weight, bone weight, body fat weight (absolute and percentage of total weight [%fat]), lean body mass (LBM). Body mass index (BMI, kg/m2) was calculated. Underweight patients were defined as having a BMI ⬍ 18.5; normal, 18.5 to 25; overweight, 25 to 30; and obese, ⬎30 kg/ m2. Using %fat, underweight women were defined as ⬍25%; overweight women, 35% to 41%; obese women, ⬎41%; underweight men, ⬍13%; overweight men, 24% to 29%; and obese men, ⬎29%.5 4. At the end of the study, a 3-day dietary history, including 1 weekend day, was obtained and analyzed for intake of energy, protein, fat, carbohydrates, vitamins, and minerals. Dietary histories containing internal inconsistencies, poor detail, or omissions were excluded. Intakes were compared with local dietary recommendations for normal individuals.6
Statistical Methods Variables were compared using Pearson product-moment correlation analysis and the Student t test. Categorical variables were compared using multiple analysis of variation.
Results A total of 140 patients were included. The diagnoses were glomerulonephritis, 40 (29%); chronic interstitial nephritis, 25 (18%); polycystic disease, 16 (11%); renovascular disease, 17 (12%); diabetic nephropathy, 15 (11%); shrunken kidneys, 22 (16%); and other, 5 (4%). The age at inclusion was 49.0 ⫾ 12.8 years, 44% were female, the height was 171 ⫾ 10 cm, and the transplant duration was 6.6 ⫾ 5.9 years. Repeat body composition studies were performed 2.9 ⫾ 0.5 years later in 115 patients. The average pred-
21
nisolone dose was 6.2 ⫾ 2.2 mg/d; azathioprine, 35 ⫾ 32 mg/d; and cyclosporine, 167 ⫾ 91 mg/d. The trough cyclosporine concentration was 131 ⫾ 55 ng/L. Dietary histories were available for 97 patients. Eighteen dietary histories were excluded because of internal inconsistencies, poor detail, or omissions. Dietary intakes are shown in Table 1. Total energy intake was lower than recommended for normal people with moderate activity (men, 10.2 to 11.4 MJ/d; women, 8.3 to 9.1 MJ/d), but physical activity was not registered in this study. Carbohydrate intake was negatively correlated with age (r ⫽ ⫺0.38, P ⬍ .001). Relative protein intake was low (⬍10%) in 2% and high (⬎15%) in 42%. Relative fat intake was low (⬍20%) in 2% and high (⬎30%) in 66%. Relative fat intake was positively correlated with age (r ⫽ 0.38, P ⬍ .001). Men had a higher proportional fat consumption (35.0% ⫾ 7.1% versus 31.6% ⫾ 7.5%, P ⬍ .03). Dietary intake of vitamin D, folic acid, iron, iodine, potassium, and selenium was below minimum recommended limits in over half of the patients. The average BMI was 25.8 ⫾ 4.3 (range, 18.1 to 39.5); men, 25.8 ⫾ 4.1; women, 25.1 ⫾ 4.9 (NS). Using BMI, 2% of patients were underweight, 44% were normal, 39% were overweight, and 14% were obese. The average %fat was 31.6 ⫾ 10.3 (range, 5.5 to 56.1); men, 27.5 ⫾ 9.4; women, 36.7 ⫾ 9.0 (P ⬍ .001). Using %fat, 13% were underweight, 21% were normal, 26% were overweight, and 39% were obese. The correlation between %fat and the inverse of BMI was linear (Fig 1), but for all patients the correlation line was above the line for normal individuals, such that at any given weight, patients had a 4% to 5% lower LBM than normal. Changes in body composition are shown in Table 2. Body composition stayed relatively stable during the study period, but a significant decrease in body fat and plasma protein and a significant increase in plasma creatinine was seen. Significant correlations with changes in body composition are shown in Table 3. Details concerning changes in bone have been reported elsewhere.7,8 Patients losing fat during the study period were characterized by initial overweight, low plasma bicarbonate, and a long transplantation duration. Although average bicarbonate levels were within normal limits, 17% of patients had a plasma bicarbonate levels below the reference
22
HEAF ET AL
Table 1. Dietary Intakes
Variable Energy (MJ/d) Energy content (kJ/100 g) Weight (kg) Protein (g) Protein (g/kg) Carbohydrates (g) Fat (g) Protein (%) Carbohydrates (%) Fat (%) Cholesterol (mg) Vitamin A (IU) Vitamin B1, thiamine (mg) Vitamin B2, riboflavin (mg) Vitamin B6, pyridoxine (mg) Vitamin B12 (g) Vitamin C (mg) Vitamin D (g) Vitamin E (IU) Vitamin K (g) Folic acid (g) Niacin (IU) Sodium (mg) Potassium (mg) Calcium (mg) Phosphate (mg) Iron (mg) Iodine (g) Selenium (g) Copper (mg) Chromium (g) Zinc (mg) Manganese (mg)
Mean ⫾ SD
Maximum-Minimum
8.33 ⫾ 2.03 485 ⫾ 108 1.70 ⫾ 0.49 71.8 ⫾ 18.5 0.97 ⫾ 0.26 17.0 ⫾ 2.3 70.8 ⫾ 25.4 15.6 ⫾ 2.8 51.0 ⫾ 6.9 33.5 ⫾ 7.2 259 ⫾ 124 1,737 ⫾ 1,472 1.15 ⫾ 0.36 1.55 ⫾ 0.61 1.63 ⫾ 0.48 5.88 ⫾ 4.31 107 ⫾ 60 3.02 ⫾ 2.87 6.86 ⫾ 3.63 70.5 ⫾ 63.4 272 ⫾ 121 24.6 ⫾ 6.8 2,804 ⫾ 794 2,645 ⫾ 817 868 ⫾ 387 1,273 ⫾ 377 10.6 ⫾ 3.1 90.5 ⫾ 65.4 29.6 ⫾ 11.9 1.13 ⫾ 0.57 29.0 ⫾ 10.4 9.90 ⫾ 2.90 3.63 ⫾ 1.71
3.20-14.09 313-868 0.63-3.49 15-121 0.3-1.8 12-23 24-149 8-23 36-69 16-48 61-656 184-7,468 0.27-1.93 0.43-2.97 0.60-3.01 1.17-27.39 20-250 0.21-13.52 1.63-17.3 1.4-351 66-988 6.6-42.3 9.5-4,614 8.8-4,607 176-1,968 353-2,151 2.6-19.8 14.6-289 5.5-63 0.24-4.11 10.3-52.7 2.08-16.71 0.96-12.21
limit (23 mmol/L), increasing to 21% at the second study. The %fat of patients with an average bicarbonate ⬍23 mmol/L decreased 2.4 ⫾ 6.6, 23 to 25 mmol/L, 1.8 ⫾ 5.4; and ⬎25 mmol/L, only 0.4 ⫾ 4.3. No significant correlations were noted with renal function or prednisolone dose. Apart from positive correlations with plasma protein, vitamin and mineral intakes did not correlate with body composition. Creatinine excretion (r ⫽ 0.61), creatinine clearance (r ⫽ 0.38), and urea excretion (r ⫽ 0.39) were highly correlated with LBM (P ⬍ .001), but not with fat weight or %fat. They did not, however, predict changes in body composition. Diabetic patients had similar body composition and dietary variables to nondiabteics, but had a significant increase in fat weight (⫹1.7 ⫾ 4.1 versus ⫺1.8 ⫾ 5.9 kg, P ⬍ .05) and %fat (⫹2.4 ⫾ 4.8 versus ⫺1.5 ⫾ 4.8%, P ⬍ .01) during the study period.
No. Patients With Normal Intake (%)
78
Normal Values6
⬍0.8
10-15 20-30 70 42 55 72 94 75 8
⬎900 ⬎1.4 ⬎1.6 ⬎1.5 ⬎2 ⬎60 ⬎5 (⬎10)
32 93
⬎300 ⬎18
17 51 96 46 17 10
⬎3,500 ⬎800 ⬎600 ⬎10 ⬎150 ⬎50
81
⬎9
Discussion Previous studies have shown a rapid increase in body weight after RT,1-4,9 partly caused by highdose steroid therapy and partly by the correction of patients’ previous anorectic uremic conditions. The patients in this study were mainly recruited some time after transplantation, and had fairly stable body composition indices over a 3-year observation period. They seemed to be adequately nourished as judged by weight, plasma albumin, and plasma protein. The study, however, revealed some problems. Although protein and energy intakes were generally sufficient, diets were characterized by an unhealthily high proportion of fat consumption, particularly among male and older patients. They were also deficient in folic acid, iron, vitamin D, vitamin B1 (thiamine), iodine, and selenium. This is in contrast to
RENAL TRANSPLANTATION AND NUTRITION
23
Figure 1. Relationship of body fat to BMI. Dotted line, normal correlation.5
the results of du Plessis et al,1 who found dietary deficiencies in iron, vitamin B6, and C, and Lacour et al,10 who found deficiency in B6. Although clinical iron deficiency is not usually present in this population, RT patients are characterized by hyperhomocysteinemia11 and accel-
erated atherosclerosis, and folic acid deficiency may well be contributing to their cardiovascular morbidity. Vitamin D levels are often deficient,7 and may contribute to posttransplantation osteodystrophy.8,12 Potassium intake was also low compared with normal; however, potassium re-
Table 2. Body Composition and Renal Function, Patients in Both Studies Only
Weight (kg) BMI (kg/m2) Bone weight (kg) Fat weight (kg) Percentage fat Lean body mass (kg) Plasma protein (g/L) Plasma albumin (mmol/L) Plasma bicarbonate (mmol/L) Plasma creatinine (mol/L) Urine creatinine (mmol/d) Creatinine clearance (mL/min) *P ⬍ .01. †P ⬍ .05. ‡P ⬍ .001.
First Study
Second Study
75.9 ⫾ 15.0 25.8 ⫾ 4.3 2.49 ⫾ 0.56 25.1 ⫾ 10.9 32.1 ⫾ 10.7 47.5 ⫾ 9.5 70.2 ⫾ 5.7 615 ⫾ 63 25.7 ⫾ 3.3 176 ⫾ 139 11.6 ⫾ 3.5 54.1 ⫾ 22.1
74.9 ⫾ 15.1 25.8 ⫾ 4.3 2.46 ⫾ 0.54 23.6 ⫾ 10.1* 31.1 ⫾ 10.0† 48.0 ⫾ 9.3 69.2 ⫾ 6.1† 584 ⫾ 60‡ 25.1 ⫾ 3.7† 220 ⫾ 215† 11.3 ⫾ 3.4 50.9 ⫾ 27.3
24
HEAF ET AL
Table 3. Significant Correlations With Changes in Body Composition Variable ⌬ Fat weight ⌬ %fat
⌬ Lean body mass
⌬ Plasma protein ⌬ Plasma albumin
Correlates
Correlation Coefficient
Significance
Initial fat weight Average plasma bicarbonate Transplantation duration Initial fat weight Initial %fat Average plasma bicarbonate Transplantation duration Initial %fat Initial lean body mass Energy Protein ⌬ Bicarbonate Vitamin K Folic acid Vitamin C Carbohydrates
⫺0.28 0.22 ⫺0.27 ⫺0.31 ⫺0.35 0.28 ⫺0.12 0.26 ⫺0.26 ⫺0.30 ⫺0.27 0.24 0.31 0.29 0.29 0.31
.03 .04 .01 .02 .01 .01 .05 .04 .04 .02 .04 .03 .04 .05 .05 .04
quirements may vary considerably in RT patients, being high in patients treated with diuretics and low in patients with poor graft function. The clinical role of low selenium and iodine intake seen in this study is unknown. Indeed, it is possible that the registered iodine deficiency is erroneous: iodine data are typically inaccurate in food analysis programs because iodine is dependent on the soil content, where and when food was harvested, etc. On the basis of these results, it would be reasonable to give patients dietary advice after RT about the benefits of reducing fat intake. Nutritional supplements with folic acid and vitamin D are advisable, and iron balance should be controlled regularly, particularly in the presence of anemia. Patients were generally overweight as judged by BMI. Obesity is atherogenic and may cause graft loss.13,14 Although the effect of BMI on morbidity and mortality is well described, it represents only a surrogate marker of obesity. Waist measurement supplies additional prognostic information, but direct measurement of fat mass by DEXA promises a major advance in our understanding of obesity. Its major disadvantage at present is that the literature concerning normative recommendations for body composition and its prognosis is limited. We compared our patients with results from Gallagher’s study of normal white US and British subjects,5 and found that at any given weight, RT patients had about 4% to 5% more body fat and correspondingly less LBM (Fig. 1). Thus, their apparent overweight
hides a condition of relative malnutrition. Using the same normative recommendations, DEXA increased the prevalence of overweight and obesity and revealed a minority group with energy malnutrition. These results suggest that BMI is insufficient in the evaluation of RT nutrition status, and should routinely be supplemented with DEXA. A slight reduction in fat weight during the study was seen. Fat weight change was negatively correlated with initial fat weight, suggesting that some patients may have taken conscious steps to reduce overweight, and vice versa. Surprisingly, steroid therapy was not correlated with change in body composition. This may also be attributable to conscious dietary changes, or simply that the prednisolone dose in this population was generally low. Similar negative findings have been found by van den Ham et al.15 The negative correlations of body composition change to initial body composition probably explain why the dietary history was of little value in predicting change, in some cases even producing paradoxical results. Two factors seemed to influence fat weight. First, long RT duration caused weight loss. This may either be a correction of the initial weight gain usually seen immediately after RT, or an expression of increasing morbidity in longterm RT patients. Second, a low or decreasing plasma bicarbonate level was associated with increased loss of both fat and LBM. As might be expected in a population with generally low renal function, the bicarbonate was in the low-normal range, and some patients were frankly acidotic.
RENAL TRANSPLANTATION AND NUTRITION
Acidosis causes anorexia and catabolism16; this study suggests that attention should be paid to correction of acid-base disturbances in this population.
References 1. du Plessis AS, Randall H, Escreet E, et al: Nutritional status of renal transplant patients. S Afr Med J 92:68-74, 2002 2. El Haggan W, Vendrely B, Chauveau P, et al: Early evolution of nutritional status and body composition after kidney transplantation. Am J Kidney Dis 40:629-637, 2002 3. Martin M, Lopes IM, Errasti P, et al: Body composition and biochemical profile as affected by diet and renal transplantation among renal patients. J Physiol Biochem 54:53-54, 1998 4. Steiger U, Lippuner K, Jensen EX, et al: Body composition and fuel metabolism after kidney grafting. Eur J Clin Invest 25:809-816, 1995 5. Gallagher D, Heymsfield SB, Heo M, et al: Healthy percentage body fat ranges: An approach for developing guidelines based on body mass index. Am J Clin Nutr 72:694-701, 2000 6. Nordiska na¨ ringsrekommendationer 1996, Nordisk Ministerråd, Nordisk Forlagshus, Nord 1996:28 7. Heaf JG, Tvedegaard E, Kanstrup IL, et al: Bone loss after renal transplantation: Role of hyperparathyroidism, acidosis, cyclosporine and systemic disease. Clin Transplant 14:457-463, 2000 8. Heaf JG, Tvedegaard E, Kanstrup IL, et al: Hyperparathy-
25
roidism and long-term bone loss after renal transplantation. Clin Transplant 17:268-274, 2003 9. Patel MG: The effect of dietary intervention on weight gains after renal transplantation. J Ren Nutr 8:137-141, 1998 10. Lacour B, Parry C, Drueke T, et al: Pyridoxal 5⬘-phosphate deficiency in uremic undialyzed, hemodialyzed, and nonuremic kidney transplant patients. Clin Chim Acta 127:205-215, 1983 11. Franke S, Muller A, Sommer M, et al: Serum levels of total homocysteine, homocysteine metabolites and of advanced glycation end-products (AGEs) in patients after renal transplantation. Clin Nephrol 59:88-97, 2003 12. Heaf JG: Bone disease after renal transplantation: Review. Transplantation 75:315-325, 2003 13. Cho YW, Terasaki PI, Cecka JM: New variables reported in the UNOS registry and their impact on cadaveric renal transplant outcomes: A preliminary study, In: , Cecka JM, Terasaki PI, (eds). Clinical Transplants 1995. Los Angeles, CA, UCLA Tissue Typing Laboratory, 1996, pp 405-415 14. Halme LM, Eklund B, Kyllonen L, et al: Is obesity still a risk factor in renal transplantation? Transplant Int 64:599-604, 1997 15. van den Ham EC, Kooman JP, Christiaans MH, et al: Relation between steroid dose, body composition and physical activity in renal transplant patients. Transplantation 69:15911598, 2000 16. Franch HA, Mitch WE: Catabolism in uremia: The impact of metabolic acidosis. J Am Soc Nephrol 9(suppl):S78-81, 1998
Dietary Habits and Nutritional Status of Renal Transplant Patients James Heaf, MD,* Ulla Jakobsen,* Erling Tvedegaard, MD,* Inge-Lis Kanstrup, MD,† and Niels Fogh-Andersen, MD‡ Background: Although dialysis nutritional problems are well described, nutritional problems after renal transplantation (RT) have received little attention. Methods: Body composition as assessed by dual-energy x-ray absorptiometry in 115 stable patients 6.6 ⫾ 5.9 years after RT and repeated 2.9 years later, when a 3-day dietary history was obtained in 79 patients. Results: Patients diet was generally sufficient, but was characterized by a high fat intake and deficiencies in folic acid, vitamin D, thiamine, iodine, selenium, and iron intake. Patients were often overweight, and at any given weight had a 4% to 5% higher proportion of body fat than normal. Loss of fat weight was related to high initial fat weight, long RT duration, and low plasma bicarbonate, but not steroid dose. Conclusion: Dietary advice concerning fat intake is indicated for RT patients, and nutritional supplements with folic acid and vitamin D are generally required. Their main nutritional problem is obesity. This is not adequately measured by body mass index, which should be supplemented by dual-energy x-ray absorptiometry. Attention should be paid to the prevention of acidosis. © 2004 by the National Kidney Foundation, Inc.
A
LTHOUGH THE NUTRITIONAL STATUS of dialysis patients is well described, few studies concerning patients after renal transplantation (RT) have been performed,1-4 and these were usually conducted in the immediate postoperative period. We decided to perform a study of stable long-term RT patients to assess dietary habits, nutritional status as judged by body composition and plasma proteins, and possible causal relationships among these.
Patients and Methods Patients A total of 135 patients with a functioning RT were included in a cross-sectional study of body *Department of Nephrology, Copenhagen University Hospital in Herlev, Herlev, Denmark. †Department of Clinical Physiology, Copenhagen University Hospital in Herlev, Herlev, Denmark. ‡Department of Clinical Biochemistry, Copenhagen University Hospital in Herlev, Herlev, Denmark. Address reprint requests to James Heaf, Graevlingestien 9, 2880 Bagsvaerd, Denmark. Email: [email protected] © 2004 by the National Kidney Foundation, Inc. 1051-2276/04/1401-0004$30.00/0 doi:10.1053/j.jrn.2003.09.005
20
composition. Three years later, 120 patients, including 5 patients not in the original study, were restudied. The patients gave informed consent in accordance to the Helsinki 2 declaration. Immunosuppressive therapy was uniform during the period of study. Transplanted recipients were treated with prednisone 1.5 mg/kg after transplantation, tapered to 20 mg daily after 20 days and 5 to 10 mg after 12 months, and with azathioprine 2 mg/kg/d tapered to 1 mg/kg/d after 20 days. Microemulsion formulation cyclosporine A (CyA) was given twice daily to achieve a trough level of 200 to 250 ng/mL (specific assay; Abbott, Abbott Park, IL), tapered to 100 to 150 ng/mL after 6 months. Antilymphocytic immunoglobulin (initially Minnesota ALG, later ATGAM) was included as induction therapy. Tacrolimus (instead of cyclosporine) and mycophenolate mofetil (instead of azathioprine) were given to occasional patients with a complicated postoperative course.
Methods The following clinical data were registered: age, sex, height, renal diagnosis, transplantation duration, and average dose of prednisone. The Journal of Renal Nutrition, Vol 14, No 1 ( January), 2004: pp 20-25
RENAL TRANSPLANTATION AND NUTRITION
following investigations were performed and were repeated 3 years later. 1. Plasma albumin, protein, creatinine, bicarbonate, urea (second analysis only), magnesium (second analysis only). 2. A 24-hour urine collection was analyzed for volume, creatinine, creatinine clearance, and urea (second analysis only). 3. The following variables were measured using dual-energy x-ray absorptiometry (DEXA; DPX-IQ, Lunar Radiation Corporation, WI): weight, bone weight, body fat weight (absolute and percentage of total weight [%fat]), lean body mass (LBM). Body mass index (BMI, kg/m2) was calculated. Underweight patients were defined as having a BMI ⬍ 18.5; normal, 18.5 to 25; overweight, 25 to 30; and obese, ⬎30 kg/ m2. Using %fat, underweight women were defined as ⬍25%; overweight women, 35% to 41%; obese women, ⬎41%; underweight men, ⬍13%; overweight men, 24% to 29%; and obese men, ⬎29%.5 4. At the end of the study, a 3-day dietary history, including 1 weekend day, was obtained and analyzed for intake of energy, protein, fat, carbohydrates, vitamins, and minerals. Dietary histories containing internal inconsistencies, poor detail, or omissions were excluded. Intakes were compared with local dietary recommendations for normal individuals.6
Statistical Methods Variables were compared using Pearson product-moment correlation analysis and the Student t test. Categorical variables were compared using multiple analysis of variation.
Results A total of 140 patients were included. The diagnoses were glomerulonephritis, 40 (29%); chronic interstitial nephritis, 25 (18%); polycystic disease, 16 (11%); renovascular disease, 17 (12%); diabetic nephropathy, 15 (11%); shrunken kidneys, 22 (16%); and other, 5 (4%). The age at inclusion was 49.0 ⫾ 12.8 years, 44% were female, the height was 171 ⫾ 10 cm, and the transplant duration was 6.6 ⫾ 5.9 years. Repeat body composition studies were performed 2.9 ⫾ 0.5 years later in 115 patients. The average pred-
21
nisolone dose was 6.2 ⫾ 2.2 mg/d; azathioprine, 35 ⫾ 32 mg/d; and cyclosporine, 167 ⫾ 91 mg/d. The trough cyclosporine concentration was 131 ⫾ 55 ng/L. Dietary histories were available for 97 patients. Eighteen dietary histories were excluded because of internal inconsistencies, poor detail, or omissions. Dietary intakes are shown in Table 1. Total energy intake was lower than recommended for normal people with moderate activity (men, 10.2 to 11.4 MJ/d; women, 8.3 to 9.1 MJ/d), but physical activity was not registered in this study. Carbohydrate intake was negatively correlated with age (r ⫽ ⫺0.38, P ⬍ .001). Relative protein intake was low (⬍10%) in 2% and high (⬎15%) in 42%. Relative fat intake was low (⬍20%) in 2% and high (⬎30%) in 66%. Relative fat intake was positively correlated with age (r ⫽ 0.38, P ⬍ .001). Men had a higher proportional fat consumption (35.0% ⫾ 7.1% versus 31.6% ⫾ 7.5%, P ⬍ .03). Dietary intake of vitamin D, folic acid, iron, iodine, potassium, and selenium was below minimum recommended limits in over half of the patients. The average BMI was 25.8 ⫾ 4.3 (range, 18.1 to 39.5); men, 25.8 ⫾ 4.1; women, 25.1 ⫾ 4.9 (NS). Using BMI, 2% of patients were underweight, 44% were normal, 39% were overweight, and 14% were obese. The average %fat was 31.6 ⫾ 10.3 (range, 5.5 to 56.1); men, 27.5 ⫾ 9.4; women, 36.7 ⫾ 9.0 (P ⬍ .001). Using %fat, 13% were underweight, 21% were normal, 26% were overweight, and 39% were obese. The correlation between %fat and the inverse of BMI was linear (Fig 1), but for all patients the correlation line was above the line for normal individuals, such that at any given weight, patients had a 4% to 5% lower LBM than normal. Changes in body composition are shown in Table 2. Body composition stayed relatively stable during the study period, but a significant decrease in body fat and plasma protein and a significant increase in plasma creatinine was seen. Significant correlations with changes in body composition are shown in Table 3. Details concerning changes in bone have been reported elsewhere.7,8 Patients losing fat during the study period were characterized by initial overweight, low plasma bicarbonate, and a long transplantation duration. Although average bicarbonate levels were within normal limits, 17% of patients had a plasma bicarbonate levels below the reference
22
HEAF ET AL
Table 1. Dietary Intakes
Variable Energy (MJ/d) Energy content (kJ/100 g) Weight (kg) Protein (g) Protein (g/kg) Carbohydrates (g) Fat (g) Protein (%) Carbohydrates (%) Fat (%) Cholesterol (mg) Vitamin A (IU) Vitamin B1, thiamine (mg) Vitamin B2, riboflavin (mg) Vitamin B6, pyridoxine (mg) Vitamin B12 (g) Vitamin C (mg) Vitamin D (g) Vitamin E (IU) Vitamin K (g) Folic acid (g) Niacin (IU) Sodium (mg) Potassium (mg) Calcium (mg) Phosphate (mg) Iron (mg) Iodine (g) Selenium (g) Copper (mg) Chromium (g) Zinc (mg) Manganese (mg)
Mean ⫾ SD
Maximum-Minimum
8.33 ⫾ 2.03 485 ⫾ 108 1.70 ⫾ 0.49 71.8 ⫾ 18.5 0.97 ⫾ 0.26 17.0 ⫾ 2.3 70.8 ⫾ 25.4 15.6 ⫾ 2.8 51.0 ⫾ 6.9 33.5 ⫾ 7.2 259 ⫾ 124 1,737 ⫾ 1,472 1.15 ⫾ 0.36 1.55 ⫾ 0.61 1.63 ⫾ 0.48 5.88 ⫾ 4.31 107 ⫾ 60 3.02 ⫾ 2.87 6.86 ⫾ 3.63 70.5 ⫾ 63.4 272 ⫾ 121 24.6 ⫾ 6.8 2,804 ⫾ 794 2,645 ⫾ 817 868 ⫾ 387 1,273 ⫾ 377 10.6 ⫾ 3.1 90.5 ⫾ 65.4 29.6 ⫾ 11.9 1.13 ⫾ 0.57 29.0 ⫾ 10.4 9.90 ⫾ 2.90 3.63 ⫾ 1.71
3.20-14.09 313-868 0.63-3.49 15-121 0.3-1.8 12-23 24-149 8-23 36-69 16-48 61-656 184-7,468 0.27-1.93 0.43-2.97 0.60-3.01 1.17-27.39 20-250 0.21-13.52 1.63-17.3 1.4-351 66-988 6.6-42.3 9.5-4,614 8.8-4,607 176-1,968 353-2,151 2.6-19.8 14.6-289 5.5-63 0.24-4.11 10.3-52.7 2.08-16.71 0.96-12.21
limit (23 mmol/L), increasing to 21% at the second study. The %fat of patients with an average bicarbonate ⬍23 mmol/L decreased 2.4 ⫾ 6.6, 23 to 25 mmol/L, 1.8 ⫾ 5.4; and ⬎25 mmol/L, only 0.4 ⫾ 4.3. No significant correlations were noted with renal function or prednisolone dose. Apart from positive correlations with plasma protein, vitamin and mineral intakes did not correlate with body composition. Creatinine excretion (r ⫽ 0.61), creatinine clearance (r ⫽ 0.38), and urea excretion (r ⫽ 0.39) were highly correlated with LBM (P ⬍ .001), but not with fat weight or %fat. They did not, however, predict changes in body composition. Diabetic patients had similar body composition and dietary variables to nondiabteics, but had a significant increase in fat weight (⫹1.7 ⫾ 4.1 versus ⫺1.8 ⫾ 5.9 kg, P ⬍ .05) and %fat (⫹2.4 ⫾ 4.8 versus ⫺1.5 ⫾ 4.8%, P ⬍ .01) during the study period.
No. Patients With Normal Intake (%)
78
Normal Values6
⬍0.8
10-15 20-30 70 42 55 72 94 75 8
⬎900 ⬎1.4 ⬎1.6 ⬎1.5 ⬎2 ⬎60 ⬎5 (⬎10)
32 93
⬎300 ⬎18
17 51 96 46 17 10
⬎3,500 ⬎800 ⬎600 ⬎10 ⬎150 ⬎50
81
⬎9
Discussion Previous studies have shown a rapid increase in body weight after RT,1-4,9 partly caused by highdose steroid therapy and partly by the correction of patients’ previous anorectic uremic conditions. The patients in this study were mainly recruited some time after transplantation, and had fairly stable body composition indices over a 3-year observation period. They seemed to be adequately nourished as judged by weight, plasma albumin, and plasma protein. The study, however, revealed some problems. Although protein and energy intakes were generally sufficient, diets were characterized by an unhealthily high proportion of fat consumption, particularly among male and older patients. They were also deficient in folic acid, iron, vitamin D, vitamin B1 (thiamine), iodine, and selenium. This is in contrast to
RENAL TRANSPLANTATION AND NUTRITION
23
Figure 1. Relationship of body fat to BMI. Dotted line, normal correlation.5
the results of du Plessis et al,1 who found dietary deficiencies in iron, vitamin B6, and C, and Lacour et al,10 who found deficiency in B6. Although clinical iron deficiency is not usually present in this population, RT patients are characterized by hyperhomocysteinemia11 and accel-
erated atherosclerosis, and folic acid deficiency may well be contributing to their cardiovascular morbidity. Vitamin D levels are often deficient,7 and may contribute to posttransplantation osteodystrophy.8,12 Potassium intake was also low compared with normal; however, potassium re-
Table 2. Body Composition and Renal Function, Patients in Both Studies Only
Weight (kg) BMI (kg/m2) Bone weight (kg) Fat weight (kg) Percentage fat Lean body mass (kg) Plasma protein (g/L) Plasma albumin (mmol/L) Plasma bicarbonate (mmol/L) Plasma creatinine (mol/L) Urine creatinine (mmol/d) Creatinine clearance (mL/min) *P ⬍ .01. †P ⬍ .05. ‡P ⬍ .001.
First Study
Second Study
75.9 ⫾ 15.0 25.8 ⫾ 4.3 2.49 ⫾ 0.56 25.1 ⫾ 10.9 32.1 ⫾ 10.7 47.5 ⫾ 9.5 70.2 ⫾ 5.7 615 ⫾ 63 25.7 ⫾ 3.3 176 ⫾ 139 11.6 ⫾ 3.5 54.1 ⫾ 22.1
74.9 ⫾ 15.1 25.8 ⫾ 4.3 2.46 ⫾ 0.54 23.6 ⫾ 10.1* 31.1 ⫾ 10.0† 48.0 ⫾ 9.3 69.2 ⫾ 6.1† 584 ⫾ 60‡ 25.1 ⫾ 3.7† 220 ⫾ 215† 11.3 ⫾ 3.4 50.9 ⫾ 27.3
24
HEAF ET AL
Table 3. Significant Correlations With Changes in Body Composition Variable ⌬ Fat weight ⌬ %fat
⌬ Lean body mass
⌬ Plasma protein ⌬ Plasma albumin
Correlates
Correlation Coefficient
Significance
Initial fat weight Average plasma bicarbonate Transplantation duration Initial fat weight Initial %fat Average plasma bicarbonate Transplantation duration Initial %fat Initial lean body mass Energy Protein ⌬ Bicarbonate Vitamin K Folic acid Vitamin C Carbohydrates
⫺0.28 0.22 ⫺0.27 ⫺0.31 ⫺0.35 0.28 ⫺0.12 0.26 ⫺0.26 ⫺0.30 ⫺0.27 0.24 0.31 0.29 0.29 0.31
.03 .04 .01 .02 .01 .01 .05 .04 .04 .02 .04 .03 .04 .05 .05 .04
quirements may vary considerably in RT patients, being high in patients treated with diuretics and low in patients with poor graft function. The clinical role of low selenium and iodine intake seen in this study is unknown. Indeed, it is possible that the registered iodine deficiency is erroneous: iodine data are typically inaccurate in food analysis programs because iodine is dependent on the soil content, where and when food was harvested, etc. On the basis of these results, it would be reasonable to give patients dietary advice after RT about the benefits of reducing fat intake. Nutritional supplements with folic acid and vitamin D are advisable, and iron balance should be controlled regularly, particularly in the presence of anemia. Patients were generally overweight as judged by BMI. Obesity is atherogenic and may cause graft loss.13,14 Although the effect of BMI on morbidity and mortality is well described, it represents only a surrogate marker of obesity. Waist measurement supplies additional prognostic information, but direct measurement of fat mass by DEXA promises a major advance in our understanding of obesity. Its major disadvantage at present is that the literature concerning normative recommendations for body composition and its prognosis is limited. We compared our patients with results from Gallagher’s study of normal white US and British subjects,5 and found that at any given weight, RT patients had about 4% to 5% more body fat and correspondingly less LBM (Fig. 1). Thus, their apparent overweight
hides a condition of relative malnutrition. Using the same normative recommendations, DEXA increased the prevalence of overweight and obesity and revealed a minority group with energy malnutrition. These results suggest that BMI is insufficient in the evaluation of RT nutrition status, and should routinely be supplemented with DEXA. A slight reduction in fat weight during the study was seen. Fat weight change was negatively correlated with initial fat weight, suggesting that some patients may have taken conscious steps to reduce overweight, and vice versa. Surprisingly, steroid therapy was not correlated with change in body composition. This may also be attributable to conscious dietary changes, or simply that the prednisolone dose in this population was generally low. Similar negative findings have been found by van den Ham et al.15 The negative correlations of body composition change to initial body composition probably explain why the dietary history was of little value in predicting change, in some cases even producing paradoxical results. Two factors seemed to influence fat weight. First, long RT duration caused weight loss. This may either be a correction of the initial weight gain usually seen immediately after RT, or an expression of increasing morbidity in longterm RT patients. Second, a low or decreasing plasma bicarbonate level was associated with increased loss of both fat and LBM. As might be expected in a population with generally low renal function, the bicarbonate was in the low-normal range, and some patients were frankly acidotic.
RENAL TRANSPLANTATION AND NUTRITION
Acidosis causes anorexia and catabolism16; this study suggests that attention should be paid to correction of acid-base disturbances in this population.
References 1. du Plessis AS, Randall H, Escreet E, et al: Nutritional status of renal transplant patients. S Afr Med J 92:68-74, 2002 2. El Haggan W, Vendrely B, Chauveau P, et al: Early evolution of nutritional status and body composition after kidney transplantation. Am J Kidney Dis 40:629-637, 2002 3. Martin M, Lopes IM, Errasti P, et al: Body composition and biochemical profile as affected by diet and renal transplantation among renal patients. J Physiol Biochem 54:53-54, 1998 4. Steiger U, Lippuner K, Jensen EX, et al: Body composition and fuel metabolism after kidney grafting. Eur J Clin Invest 25:809-816, 1995 5. Gallagher D, Heymsfield SB, Heo M, et al: Healthy percentage body fat ranges: An approach for developing guidelines based on body mass index. Am J Clin Nutr 72:694-701, 2000 6. Nordiska na¨ ringsrekommendationer 1996, Nordisk Ministerråd, Nordisk Forlagshus, Nord 1996:28 7. Heaf JG, Tvedegaard E, Kanstrup IL, et al: Bone loss after renal transplantation: Role of hyperparathyroidism, acidosis, cyclosporine and systemic disease. Clin Transplant 14:457-463, 2000 8. Heaf JG, Tvedegaard E, Kanstrup IL, et al: Hyperparathy-
25
roidism and long-term bone loss after renal transplantation. Clin Transplant 17:268-274, 2003 9. Patel MG: The effect of dietary intervention on weight gains after renal transplantation. J Ren Nutr 8:137-141, 1998 10. Lacour B, Parry C, Drueke T, et al: Pyridoxal 5⬘-phosphate deficiency in uremic undialyzed, hemodialyzed, and nonuremic kidney transplant patients. Clin Chim Acta 127:205-215, 1983 11. Franke S, Muller A, Sommer M, et al: Serum levels of total homocysteine, homocysteine metabolites and of advanced glycation end-products (AGEs) in patients after renal transplantation. Clin Nephrol 59:88-97, 2003 12. Heaf JG: Bone disease after renal transplantation: Review. Transplantation 75:315-325, 2003 13. Cho YW, Terasaki PI, Cecka JM: New variables reported in the UNOS registry and their impact on cadaveric renal transplant outcomes: A preliminary study, In: , Cecka JM, Terasaki PI, (eds). Clinical Transplants 1995. Los Angeles, CA, UCLA Tissue Typing Laboratory, 1996, pp 405-415 14. Halme LM, Eklund B, Kyllonen L, et al: Is obesity still a risk factor in renal transplantation? Transplant Int 64:599-604, 1997 15. van den Ham EC, Kooman JP, Christiaans MH, et al: Relation between steroid dose, body composition and physical activity in renal transplant patients. Transplantation 69:15911598, 2000 16. Franch HA, Mitch WE: Catabolism in uremia: The impact of metabolic acidosis. J Am Soc Nephrol 9(suppl):S78-81, 1998