**Estimating Urea Volume in Amputees on Peritoneal Dialysis by Modified Anthropometric F ormulas**

**Antonios H. Tzamaloukas,' Glen H. Murata2**

*BodY composition determines bot!Y water content (the fraction bot!Y water/bot!Y weight). With developing obesity, bot!Y weight and body water increase, but body water content decreases. The anthropometric formulas for urea volume (bot!Y water) for Kt/V computations in nonamputated peritoneal dialysis subjects rejlect this fundamental rule ofbot!Y composition. How. evel; the use of uncon-ected anthropometric formulas in amputees provides bot!Y water content estimates inconsistent with the estimates of bot!Y composition obtained from nutritional assessment. Con-ected estimates of urea volume can be obtained in three steps: (1) The non-amputated weight at the same bot!Y composition is computed by dividing the weight at the urea kinetic study (postamputation) by (1 -the fractional weight loss from the amputation); (2) bot!Y water and bot!Y water content at this nonamputated weight are obtailK!dfrom the appropriate anthropometric formula; (3) at the time of the urea kinetic study, post-amputation, bot!Y water is equal to the estimate ofbot!Y water content obtained from step 2 times the body weight at the urea kinetic study. The con-ected estimates of urea volume provide bot!Y water content values agreeing with the estimates from nutritional assessment.*
**Key words**

Amputation, body composition, body water, urea clearance, urea volume
**From:**

Sections of Nephrology' and General Internal Medicine,2 Albuquerque Veterans Affairs Medical Center and the University of New Mexico School ofMedicine, Albuquerque, New Mexico, U.S.A.

**Introduction**

Urea volume (V) for calculating KtlV in peritoneal dialysis can be estimated by several methods. Among these methods, the anthropometric formulas have definite advantages including ease of calculation and low cost. More importantly, the anthropometric formulas allow specific corrections for conditions different &om those present during their derivation (I ). The Watson anthropometric formula (2) have been proposed as the method of choice for estimating V (1,3). Amputation is one of the conditions not encountered in the populations from which the anthropometric formulas were derived. Nevertheless, amputation of one or more extremities is encountered in peritoneal dialysis patients suffering from diabetes or peripheral vascular disease. Correction of the anthropometric formulas for amputation is feasible (4). Here we present the rationale for, and an exampIe of, this correction.
**Rationale**

This section assumes that patients are at their dry weight. The main determinant of body water content (defined as the fraction body water/body weight) at dry weight in the anthropometric formulas is body composition (5,6). In nutritional assessments, one of the ways of estimating body composition is by computing the fractional ( or percent) deviation of body weight from ideal (FDW). The formula for FDW is: actual weight ideal weight)/ideal weight (5). In the same person, body composition is the sole determinant of body water content. Figure I shows body weight, body water estimated by the Watson formula, and body water content in a 59-year-old man, 183 cm tall, if his body composition were to change between extreme malnutrition (FDW = -0.5) and pronounced obesity (FDW = 0.5). The figure shows that as progressive obesity develops, body water increases but body water content decreases. This finding is consistent with the fact that the water content of fat-free body mass is approximately 72%, while fat tisue has a much lower water content. Figure 1 also shows that a unique value of body water content corresponds to each FDW in the same subject and year. Therefore, if the body water content is known, body composition (the FDW) 'can be estimated. The principle that will be explored in this report is that water content is the same in an amputated extremity and the remaining body. An appropriate correction of ideal weight, and consequently of FDW, for amputation is routinely used in nutritional assessments (7). Body composition estimates from anthropometric estimates ofV (Figure 1) and from nutritional assessments should be in agreement.

**Proposed correction**

*Example*

A 59-year-old man developed traumatic end-stage renal failure after a motor vehicle accident, which also resulted in amputation ofboth lower extremities, almost at the hips. He had a urea kinetic study two months after commencing continuous ambulatory peritoneal dialysis (CAPD). His height was 183 cm before the accident. At the time of the study his body length was 107 cm and his weight was 50.1 kg. Raw urea clearance was 65 L weekly. Nutritional assessment for an amputee (7) disclosed an ideal weight of 51.1 kg and FDW = -0.019.

Kt/V was calculated first with uncorrected Watson estimates of V, using the age of 59 years, 50.1 kg for weight, and two values for height: ( 1 ) for a height of 107 cm, V = 25.2 L and Kt/V = 65/ 25.2 = 2.58 weekly. Body water content would be 25.2/50.1 = 0.502 L/kg. This water content is consistent with obesity (FDW = +0.231 ); (2) for a height of 183 cm, V = 33.3 L, Kt/V = 1.95 weekly, and body water content = 0.665 L/kg, which is consistent with wasting (FDW = -0.382). Neither of the two uncorrected estimates of V is consistent with the patient's body composition. In addition, the two estimates of Kt/V vary by approximately 33%.

Corrected estimates of V and Kt/V consistent with the patient's body composition are obtained in three steps.
*Step 1: Determination of the hypothetical nonamputated weight at the same body composition*

The percent of body weight lost to amputations is found in nomograms. Like the anthropometric formulas, these nomograms present average estimates of the fraction of body weight represented by each body part, when weight and composition of this part are not distorted by disease. Also like the anthropometric formulas, the nomograms allow the calculation of the weight of the body part when there is a global change in body composition, such as obesity. Figure 2 reproduces one published nomo-gram (8). From this figure, the amputated legs account for 37% of the weight of the subject studied. Therefore, the weight of 50.1 kg represents 63% of the nonamputated weight at the same body composition, and the non amputated weight would be 50.1/0.63 = 79.5 kg. Note: Using, as has been suggested, the weight of the amputated extremity, ifknown, to reconstruct body weight is erroneous for two reasons: ( 1) body composition usually changes between the time of the amputation and the time of the urea kinetic study (4). The change in body composition would have affected the weight of the amputated extremity; and (2) the composition of the amputated extremity was probably different from that of the other body parts at the time of the amputation. This altered composition would also have unpredictably affected the weight of the amputated extremity even if body composition had not changed between amputation and the urea kinetic study. In any event, the only relevance of the amputation on Kt/V is through its effect of the calculation ofV. The water content of the amputated part is not part of this V.

*Step 2: Determination ofbody water at the hypothetical nonamputated weight*

The Watson formula for a 59-year-old man with a height of 183 cm and a weight of 79.5 kg reveals V = 43.2 L, and body water content = 43.2/79.5 = 0.544 L/kg. This water content is consistent with the patient's nutritional status (FDW = -{1.019).
*Step 3: Determination ofbody water of the amputated subject*

Based on the principle presented in the Rationale section, body water content is 0.544 L/kg and therefore body water is 0.544'50.1 = 27.3 L. Corrected KtlV is 65/27.3 = 2.38 weekly.
**Discussion**

The correction of the Watson formula proposed in this report is consistent with the effect ofbody composition on body water content (Figure I ). This correction should be considered only as a reasonable approximation, since small variations in water content between the extremities and other body parts are probable ( 4). The accuracy of the corrected esti mates ofV can be assessed only by comparing these estimates to direct measurements of body water or to estimates of V obtained by urea kinetics during hemodialysis sessions (6). Nevertheless, uncorrected anthropometric estimates ofV clearly disagree with estimates of body composition and provide conflicting results in bilateral amputees depending on the height estimate used in the anthropometric formula.

The precision of the corrected estimates can improve by careful evaluation of the percent of body weight lost to amputation, reliable knowledge of the preamputation height in bilateral leg amputees, and careful assessment of dry weight. The Watson estimates ofV are more or less precise in subjects studied at their dry weight (I). Subjects with either volume excess or volume deficit, whether they have amputations or not, require a special correction of the Watson estimates of V (9). In amputated subjects with a superimposed volume disorder, the correction for the amputation should be performed using their dry weight. Then the difference between actual weight and dry weight should be added to (volume excess) or subtracted from (volume deficit) the dry weight V estimate. This sum constitutes the V entered in the KtlV formula.
**References**

- Tzamaloukas AH, Murata GH. Estimating urea volume in peritoneal dialysis: pitfalls and corrections. Int J ArtifOrgans (in press).
- Watson PE, Watson ID, Batt RD. Total body water volumes for adult males and females estimated from simple anthropometric measurements. Am J Clin Nutr 1980; 33:27-39.
- Tzamaloukas AH, Murata GH. Computational formulas for clearance indices in continuous ambulatory peritoneal dialysis. Perit Diallnt 1996; 16:13-14.
- Tzamaloukas AH, Saddler MC, Murphy G, et a/. Volume of distribution and fractional clearance of urea in amputees on continuous ambulatory peritoneal dialysis. Perit Dial Int 1994; 14:35~1.
- Tzamaloukas AH, Murata GH, Malhotra D, et a/. Urea kinetic modeling in continuous peritoneal dialysis patients. Effect of body composition on the methods computing urea volume of distribution. ASAIO J 1993; 39:M359-62.
- Dumler F, Cruz C. The method used for volume estimation significantly influences KprT/V results in peritoneal dialysis patients. In: Khanna R, ed. Advances in peritoneal dialysis. Toronto: Peritoneal Dialysis Publications Inc., 1995; II :88-92.
- Stratman D, Donnelly LT. Determination ofideal body weight and nutritional requirements postamputation. Orthop Nurs 1984; 3:37-40.
- Hopkins B. Assessment of nutritional status. In: Gottschlich MM, Matarese LE, Schronts EP, eds. Nutrition support dietetics core curriculum, 2nd ed. Silver Spring. Maryland: American Society for Parenteral and Enteral Nutrition, 1993: 15 70.
- Tzamaloukas AH. Effect of edema on urea kinetic studies in peritoneal dialysis patients. Perit Dial Int 1994; 14:398-401.
- Brunstrom MA. Clinical kinesiology, 3rd ed. Philadelphia: FA Davis, 1981.

*Corresponding author :*

Antonios H. Tzamaloukas, MD, Renal Section (1IIC), VA Medical Center, 2100 Ridgecrest Drive, SE, Albuquerque, New Mexico 87108 U.S.A