V underestimates the hemodialysis dose in women and small men

original article

http://www.kidney-international.org & 2008 International Society of Nephrology

see commentary on page 262

Kt/V underestimates the hemodialysis dose in women and small men Elaine M. Spalding1, Shahid M. Chandna1, Andrew Davenport2 and Ken Farrington1 1

Renal Unit, Lister Hospital, Coreys Mill Lane, Stevenage, Herts, UK and 2Renal Unit, Royal Free Hospital, Pond St, London, UK

Current guidelines suggest a minimum Kt/V of 1.2 for three weekly hemodialysis sessions; however, using V as a normalizing factor has been questioned. Parameters such as weight0.67 (W0.67) and body surface area (BSA) that reflect the metabolic rate may be preferable. To determine this, we studied 328 hemodialysis patients (221 male) with a target Kt/V of 1.2. Using this relationship and the individual’s Watson Volume, we calculated the Kt, Kt/BSA, and Kt/W0.67 equivalent to the target and measured the effects of body size and gender on these parameters for each patient. The target corresponded to a range of equivalent Kt/BSA and Kt/W0.67 each significantly higher in males than females and in larger than smaller males. V/BSA and V/W0.67, the conversion factors of Kt/V to Kt/BSA and Kt/W0.67 respectively, were significantly greater in males than females and heavier than lighter men. Our study shows that if Kt/BSA and Kt/W0.67 reflect the true required dose, prescribing a target Kt/V of 1.2 would underestimate this in females and in small males. Further work is required to develop clinical outcome-based adequacy targets. Kidney International (2008) 74, 348–355; doi:10.1038/ki.2008.185; published online 28 May 2008 KEYWORDS: hemodialysis; adequacy; Kt/V; metabolic size; body surface area; total body water

Correspondence: Ken Farrington, Renal Unit, Lister Hospital, Coreys Mill Lane, Stevenage, Herts SG1 4AB, UK. E-mail: [email protected] Received 2 December 2006; revised 7 February 2008; accepted 26 February 2008; published online 28 May 2008 348

Gotch and Sargent’s re-analysis of the National Cooperative Dialysis Study established the dimensionless parameter Kt/V as an index of hemodialysis (HD) adequacy, where K was dialyzer urea clearance, V urea distribution volume equating to total body water, and t duration of the HD session.1 This parameter, derived from single-pool urea kinetic modeling, has become the gold standard for HD prescription and dose monitoring. Modifications to take account of factors such as urea generation and two-pool effects have become well accepted,2,3 and the concept of Kt/V, as a marker of adequacy, has been extended to other modalities, including peritoneal dialysis.4 V is an independent predictor of survival across the range of delivered HD dose, an effect that has been attributed to malnutrition in a poor-risk group of patients with low body mass.5,6 Such findings have fueled criticism of the use of V as the normalizing factor for Kt.7 Furthermore, there were suggestions from the HEMO Study that women in the intervention group (achieved two-pool Kt/V 1.53) had a significantly lower mortality than those in the standard dose group (achieved two-pool Kt/V 1.16).8 Though differences in body size could not be implicated,9 the suspicion of a denominator effect remains. In prescribing HD, K is estimated from the urea mass transfer coefficient (KoA) of the dialyzer together with the anticipated blood and dialysis fluid flow rates. V is usually estimated from anthropometric equations such as the Watson formulae,10 though use of kinetic-based estimates may be more desirable. In addition, the Watson V is often used to estimate HD efficiency, a comparison of prescribed and delivered dose. Estimating Watson V in male and female subjects utilizes different equations that reflect known differences in body composition. For any given age, height, and weight, the calculated V is lower in women than in men. Furthermore, the formulae for Watson V were based on data from 30 different anthropometric studies selected to be representative of a healthy Western population. The relevance of these formulae to an aged, multiethnic, highly comorbid HD population in which malnutrition and volume overload are prevalent is perhaps questionable. A further problem arises from the requirement for dialysis to replicate renal excretion of metabolic waste products. It might be expected that patients with higher metabolic rates would require more dialysis than those with lower rates. Kidney International (2008) 74, 348–355

original article

EM Spalding et al.: Kt/V underestimates hemodialysis dose

Hence factors reflecting metabolic rate (‘metabolic size’) might be more appropriate normalizing factors. It has been suggested that the strongest single correlate of resting energy expenditure (REE) is fat-free mass,11 and although there are now simple non-invasive measurement techniques, use of this parameter as a normalizing factor would require its serial prospective measurement.12 Developing on this theme, it has been suggested that the magnitude of the ratio of body cell mass to fat-free mass is a major determinant of whole-body REE.13 Recent work, although, has highlighted the contribution of the visceral organ mass, estimated by whole-body magnetic resonance imaging, to resting energy metabolism and uremic toxin production.14 There are simpler estimates of metabolic size. Basal metabolism (P) is related to body mass (W, kg) by the expression: P ¼ aW b

ð1Þ

Where a ¼ mass coefficient, which within a species is a constant, b ¼ mass index, which within species approximates to 0.67–0.7515–17 Body surface area (BSA) is also related to resting metabolic rate, and correlates with body composition in both sexes.18 Normalizing glomerular filtration rate to BSA is standard practice. Normalizing dialyzer clearance in the same way, would seem to be a logical extrapolation. A common feature of the anthropometric formulae for estimation of BSA, which is also a feature in Equation 1, is the incorporation of body mass raised to an index, the value of which is less than unity. Use of such parameters as the normalizing factor has the effect of reducing the relative importance of the denominator as body mass increases. Patients with low body mass would require more dialysis per unit mass than their heavier counterparts. This concept has received support from groups deploying empirical and theoretical approaches. Lowrie et al.7,19,20, using retrospective registry data, investigated the relationships between outcome, both derived and measured Kt, and a range of estimates of body size. They found that increasing values of dialysis dose and body size, however estimated, were both associated with reduced mortality. Furthermore, Kt/BSA, unlike single-pool and equilibrated Kt/V, was significantly associated with death risk; patients with a low Kt/BSA having an increased hazard ratio.19 A theoretical basis for these and similar observations has been suggested and supported by the finding of a proportionately larger visceral organ mass in smaller than in larger patients.14,21 Hence REE and uremic toxin production may be proportionately greater in smaller patients leading to the requirement of a proportionately higher dialysis dose. This may be compounded by their smaller muscle mass, and hence their smaller volume of distribution for soluble uremic toxins and by their smaller fat mass and hence their smaller volume of distribution for fat-soluble uremic toxins. On all these counts, smaller patients would be more susceptible to the toxic effects of uremia. Kidney International (2008) 74, 348–355

Table 1 | Demographic and biochemical parameters of study population

Number Age (years) Weight (kg) Height (m) Watson volume (l) Body surface area (m2) % Caucasian % Diabetes Hemoglobin (g per 100 ml) Serum albumin (g/l)

All patients

Males

Females

P-value

328 62.5±14.9 71.3±15.9 1.69±1.06 37.3±7.1 1.83±0.24 80.5 20.5 11.5±1.5

221 62.3±15.1 75.2±15.3 1.74±0.81 40.5±5.9 1.91±0.221 82.3 22.6 11.5±1.5

107 62.7±14.5 63.1±14.0 1.59±0.74 30.6±3.8 1.67±0.21 76.6 15.0 11.4±1.5

NS o0.001 o0.001 o0.001 o0.001 NS NS NS

33.7±5.2

33.9±5.3

33.4±4.7

NS

NS, not significant.

This study was undertaken to evaluate the effect of substituting BSA and W b for V as denominator in Kt/V, on the distribution of target dialysis dose within our HD population. We have generally considered b to be 0.67. The use of V 0.67 as the normalizing factor for the in-center HD arm of the ongoing NIH Frequent Haemodialysis Study lends some justification for this choice.22 Given a target Kt/V of 1.2, we sought to determine the equivalent HD dose expressed in terms of Kt/BSA and Kt/W 0.67 and to determine the effect of gender and body size on these relationships. RESULTS Patients

There were 328 patients, 221 men (age 62.3±15.1 years) and 107 women (age 62.7±14.5 years). Dialysis vintage was 1–253 months (mean 44.5±40.1). Mean anthropometric and biochemical parameters are shown in Table 1. Relationship between Watson volume, BSA, W 0.67, and weight

The relationship between Watson volume and weight (Figure 1) was best approximated by separate straight lines for men (R2 ¼ 0.855) and women (R2 ¼ 0.960), which do not pass through the origin and which diverge as weight increases. The relationships between weight and BSA were best approximated as a power function of the form y ¼ aWb, where a is a constant and b is a constant, the value of which is less than unity. There is a small gender difference (a ¼ 0.16, b ¼ 0.575, R2 ¼ 0.979 (men) and a ¼ 0.16, b ¼ 0.568, R2 ¼ 0.982 (women)). The relationship between weight and Wb is depicted in Figure 2 in a series of curves for values of b between 0 and 1 (clearly a ¼ 1 and R2 ¼ 1 for all cases, and there is no gender differences). Variation of V/BSA and V/W 0.67with gender and body weight

The quantities V/BSA and V/W 0.67 can be considered as conversion factors of Kt/V to Kt/BSA and Kt/V 0.67 respectively. Both ‘conversion factors’ were significantly greater in men than in women (by 16 and 17.4% respectively; Po0.001 in both cases) and greater in larger (above mean V for men) than in smaller men (by 6 and 3.7% respectively; Po0.001 in 349

original article

EM Spalding et al.: Kt/V underestimates hemodialysis dose

70 20

60

∗

15 V/ BSA

Watson V (l)

50

∗

40

10

30 5

20 10

0

0

∗

20

40 60 80 100 120 140 Dry weight (kg)

2.0

Figure 1 | Relationship between Watson volume (V) and dry weight in 328 HD patients. The lines displayed represent best linear regression fit in men and women. HD, hemodialysis.

V/W 0.67

0

∗

1.5 1.0 0.5

[Body weight] b

120 100 0.9

80 60

0.8

40

0.7 0.67 0.6 0.5 0.1

20 0 0

20

40

Values of exponent b

1

140

60 80 100 120 140 Body weight (kg)

Figure 2 | Relationship between Wb and dry weight. Separate non-linear regression lines (of the form y ¼ aWb) are shown for a ¼ 1, b ¼ 0.1–1.0, at increments of 0.1. For details see text. Bold lines represent the curves for b ¼ 0.67 and b ¼ 1. For clarity data points omitted.

both cases) (Figure 3). Smaller women (less than mean women V) tended to have slightly higher values of both parameters (0.5 and 3.5% respectively) than larger women (significant for V/W 0.67; Po0.007).

0 Small Large Small female female male

Large male

Figure 3 | Relationship between Watson volume (V), body surface area, and W 0.67 in 328 HD patients. Upper panel: relationship between Watson V and body surface area in 328 HD patients clustered into four groups—small women (omean V for women), large women (4mean V for women), small men (omean V for men), and large men (4 mean V for men). *Po0.001 for comparison with all other groups. Lower panel: relationship between Watson V and W0.67 in 328 HD patients clustered into four groups—small women (omean V for women), large women (4mean V for women), small men (omean V for men), and large men (4mean V for men). *Po0.001 for comparison with all other groups. Po0.007 for comparison between small and large women. HD, hemodialysis.

slightly but significantly greater in smaller than in large women. The differences between mean Kt/BSA and Kt/W 0.67 in these four gender–body size groups and their degrees of significance mirrored those described above for V/BSA and V/W 0.67.

Adequacy parameters equivalent to Kt/V of 1.2

The values of the adequacy parameters Kt, Kt/BSA, and Kt/V 0.67, equivalent to a prescribed Kt/V of 1.2, are shown in Figure 4 and Table 2. All parameters were bimodally distributed with separate peaks for men and women (Figure 4), and all were significantly greater in men (Table 2). All were greater in patients in the upper tertile of body size (V) than in those in the middle tertile. Likewise, those in middle tertile were greater than in those in the lower (Table 3). While all equivalent adequacy parameters were significantly greater in larger (for definitions see above) than in smaller men, only Kt was greater in larger than in smaller women. Kt/BSA in women was independent of body size, while Kt/W 0.67 was 350

Body size, gender, and estimated ‘required’ target Kt

Figure 5 depicts the relationship in individual hemodialysis (HD) patients between body size (Watson V) and two values of ‘required’ target Kt, estimated as functions of BSA according to the work of Lowrie et al.20 Figure 5A demonstrates this relationship for the ‘double-reciprocal’ estimate and Figure 5B for the multiplicative estimate (see Materials and Methods section and the figure legend for detail). For both estimates, over the whole range of encountered V, ‘required’ target Kt increases in a curvilinear fashion with increasing V. There are also stark differences between men and women with respect to their ‘required’ Kidney International (2008) 74, 348–355

original article

Number of patients

EM Spalding et al.: Kt/V underestimates hemodialysis dose

40

Target Kt/BSA and Kt/W0.67 values equivalent to a range of target Kt/V values

30

Mean equivalent Kt/BSA and Kt/W0.67 for target Kt/V between 0.7 and 1.7 are shown in Figure 6 in relation to gender and body size. For all target Kt/V, mean equivalent Kt/BSA and Kt/W0.67 levels were significantly greater in men than in women, the differences increasing with target Kt/V. They were also significantly greater in large than in small men, the differences increasing with target Kt/V. In women, there was no difference between equivalent Kt/BSA in large and small individuals, while equivalent Kt/W0.67 was consistently slightly greater in smaller women. The effect was small but increased with target Kt/V. These differences again mirrored those described above for V/BSA and V/W0.67.

20 10 0

Number of patients

20

30

40

50 60 Kt (l)

70

80

90

80 70 60 50 40 30 20 10 0

Range of Kt/Wb equivalent to Kt/V of 1.2 15

20

25

30

35

Kt/BSA (l/m2) Number of patients

50 40 30 20 10 0 1.5

2.0

2.5 3.0 Kt/W 0.67

3.5

4.0

Figure 4 | Histograms of distribution by gender of Kt (upper panel), Kt/BSA (middle panel), and Kt/W0.67 values (lower panel) equivalent to target Kt/V of 1.2 in 328 hemodialysis patients.

Table 2 | Adequacy parameters equivalent to prescribed Kt/V of 1.2, by gender Parameter Kt (l) Kt/BSA (l/m2) Kt/W0.67 (l/kg0.67)

Whole group

Male

Female

P-value

44.7±8.4 24.3±2.0 2.57±0.23

48.7±7.1 25.5±1.4 2.70±0.15

36.7±4.5 21.9±0.3 2.30±0.09

o0.001 o0.001 o0.001

Table 3 | Adequacy parameters equivalent to prescribed Kt/V of 1.2, by tertiles of body size (V) Tertiles of V Parameter Kt (l) Kt/BSA (l/m2) Kt/W0.67 (l/kg0.67)

Lower

Middle

Upper

P-value

35.7±3.2 22.4±1.2 2.38±0.17

44.5±2.2 24.4±1.5 2.59±0.21

54.1±5.6 26.2±1.1 2.75±0.15

o0.001 o0.001 o0.001

target Kt. In very small patients, values of ‘required’ target Kt are similar in women and men. However, in all but these very small individuals, the required target is greater in women and the gender difference increases with increasing body size. Kidney International (2008) 74, 348–355

Figure 7 shows the relationship between the values of Kt/Wb equivalent to a Kt/V of 1.2 and corresponding values of Wb, for values of b between 0 and 1. On right side of each panel as W1 ¼ W, Kt/Wb ¼ Kt/W, which is directly proportional to Kt/V. On the left side of each panel, as W0 ¼ 1, Kt/Wb ¼ Kt. The upper panel demonstrates the effect of gender on this relationship. Mean Kt/Wb in women is expressed as a proportion of the value in men. When b ¼ 1, equivalent Kt/W is 10% less in women than in men. As b decreases, the gender difference in Kt/Wb increases, so that at b ¼ 0, mean equivalent Kt was 24.6% lower in women. The middle panel depicts Kt/Wb in relation to Wb in the three tertiles of body size (V), expressed as a percentage of values in the middle tertile. As b decreases, equivalent Kt/Wb increases in the upper tertile and decreases in the lower tertile. The lower panel depicts the combined effects of gender and size on equivalent Kt/Wb. The relationship is more complex. The dashed vertical lines in each panel represent the values of b, 0.5378 (the exponent of weight in the Haycock BSA) and 0.67. At these levels of weight exponent, both gender and body size have a potentially major impact on target HD adequacy expressed as Kt/BSA or Kt/W0.67. DISCUSSION

More than average weight patients have a survival advantage.5,23,24 This has been attributed to an excess mortality in those with low body mass index, due to malnutrition and cachexia,5,6 though the effect persists after adjustment for comorbidity, diabetes, and morbid obesity.24 Our study suggests that other factors may contribute to the excess mortality in smaller patients. If the appropriate normalizing factor for the required dialysis dose were metabolic size (BSA or W 0.67) rather than body mass (W), the relative importance of the normalizing factor in determining the target HD dose (Kt/BSA or Kt/W 0.67) would be greater in smaller individuals. Hence use of V as the normalizing factor in prescribing dialysis would tend to result in inadequate treatment for smaller patients, as has been previously suggested.7,19,20 Recent observations suggesting that smaller patients may have disproportionately higher rates of uremic toxin 351

original article

70

70 'Required' Kt - 'Multiplicative' (l)

'Required' Kt - 'Double Reciprocal' (l)

EM Spalding et al.: Kt/V underestimates hemodialysis dose

60 50 40 30

Male Female

60 50 40 30 20

20 20

30

40 50 60 Watson V (l)

70

80

20

30

40 50 60 Watson V (l)

70

80

Figure 5 | Plots of estimates of ‘required’ Kt against Watson V. For definitions of ‘double reciprocal’ and ‘multiplicative,’ see reference Lowrie et al.20 and Materials and Methods section. Curves were fitted using the LOESS locally weighted regression technique. Continuous lines represent relationship in men. Broken lines represent relationship in women.

35 30

4 M > mean V M < mean V F > mean V F < mean V

25 20

3.5 Equivalent Kt/W 0.67

Equivalent Kt/BSA (l/m2)

40

3

M > mean V M < mean V F > mean V F < mean V

2.5 2

15

1.5

10

1

0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7

0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7

Prescribed Kt/V

Prescribed Kt/V

Figure 6 | Equivalent Kt/BSA and Kt/W0.67 for a range of prescribed Kt/V in 328 HD patients. Left panel: equivalent Kt/BSA for a range of prescribed Kt/V in 328 HD patients clustered into four groups—small women (omean V for women), large women (4mean V for women), small men (omean V for men), and large men (4mean V for men). Right panel: equivalent Kt/W0.67 for a range of prescribed Kt/V in 328 HD patients clustered into four groups—small women (omean V for women), large women (4mean V for women), small men (omean V for men), and large men (4mean V for men). BSA, body surface area; HD, hemodialysis.

production, coupled with lower volumes of distribution, than their larger peers,14,21 provide a strong theoretical basis for the use of an alternative denominator in dialysis prescription. This denominator should reflect the rate of production of uremic toxins, and not just their volume of distribution. Our use of BSA and W0.67 may represent a step toward this goal. Previous empirical analysis has clearly demonstrated a strong curvilinear relationship between outcome-defined target Kt and BSA.20 Although best described statistically by a double-reciprocal relationship (KtDR; see Equation 5), the relationship can also be described by a power function (KtM ¼ 34.6  BSA0.63; see Equation 6). This empirically derived expression is similar to our starting premise, which was based on the hypothetical relationship between target Kt and P, which can be represented as a power function of weight (P ¼ aWb: from Equation 1, where we have generally taken b ¼ 0.67). Though these exponents are similar, it 352

should be noted that BSA is also related to W 0.54 (in the Haycock formula), hence KtM is approximately related to W 0.36. Our use of Wb and BSA as alternative denominators (to V) for dialysis dose has allowed us to demonstrate that patients of different body size and gender, dialyzed to the same target Kt/V, may be receiving significantly different treatment doses, when these are quantified by other means. What we have not done is to define a putative target dialysis dose expressed in terms of Kt/BSA or Kt/Wb. This would need to be determined by rigorous analysis of outcome data, as has been previously undertaken in the evaluation of target KtDR.25 Previous work has also shown that low levels of Kt/BSA may be associated with increased mortality risk.19 Extrapolating from Figure 5, Lowrie et al.19 would suggest an increased hazard ratio when Kt/BSA o23 l/m2. The same figure suggests that high levels of Kt/BSA may also confer a slightly increased mortality risk,19 though this is less clear and further work is required on this issue. Kidney International (2008) 74, 348–355

EM Spalding et al.: Kt/V underestimates hemodialysis dose

Equivalent Kt/W b Relatie to males

Gender Male

100

Female

90 80 70

Equivalent Kt/W b Relatie to middle V tertile

Body size: Tertiles of V 120 Lower tertile of V

110 100 90 80 70

Middle tertile of V Upper tertile of V

Equivalent Kt/W b Relatie to males > mean V

Gender and body Size 110 100 90 80 70 60

Female mean V Male < mean V Male > mean V

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Weight expontent (b)

Figure 7 | Kt/Wb equivalent to Kt/V ¼ 1.2 for a range of b between 0 and 1 in 328 HD patients. In the upper panel, mean values for female patients are shown relative to mean values in men. In the middle panel, mean values are shown in tertiles of V, relative to mean values in medium tertile. In the lower panel, mean values are shown in four groups (small women (omean V for women), large women (4mean V for women), small men (omean V for men), and large men (4mean V for men), relative to the values in large men). The dashed vertical lines in each panel represent the values of b, 0.5378 and 0.67, respectively. The former being the power of weight used in the Haycock formula for BSA, and the latter, an approximation to ‘metabolic size’.BSA, body surface area; HD, hemodialysis.

Body size is clearly important in determining required dialysis dose, but alone does not explain the bimodal distribution observed in our plots of equivalent Kt/BSA and Kt/W0.67. These point to a specific effect of gender and reflect the different relationship between V and BSA in men and women. This is also apparent from our plots of ‘required’ target Kt against V (Figure 5), which suggest that for a given level of V, the required target dialysis dose is greater in women than in men. These gender differences relate to an increased adipose tissue-to-fat-free mass ratio in women, who therefore have disproportionately low V for their body size. The different formulae for Watson V for men and women reflect this and predict that for any given age, height, and weight, calculated V is lower in women than in men. It may be that these formulae do not accurately reflect total body water in uremic men or women or both, and that the bimodal distribution is artefactual. In this case, use of a kinetic V in dialysis prescription would be more appropriate. If, as is likely, Watson V does reflect total body water in this population, then the bimodal distribution represents a true gender difference, and the same concerns would arise even Kidney International (2008) 74, 348–355

original article

with use of a kinetic V in dialysis prescription. It seems that visceral mass, which reflects the high metabolic rate compartment, may have a similar relationship to body size in men and women.14 Hence, it may be that effects of body size and gender on required dialysis dose have different pathophysiological bases, the influence of body size being mediated through its non-linear relationship with uremic toxin production, and that of gender, through gender-related differences in uremic toxin distribution volume. Use of BSA or W0.67 as the normalizing factor for HD prescription may represent a more physiological starting point, but many of the issues discussed above demonstrate the need to refine and validate the concept before it can be used in the clinical arena. There are other potential problems. BSA or W 0.67 relate only to basal metabolic requirements, while the generation of metabolic waste must relate to total energy expenditure. Total energy expenditure consists of REE plus physical activity energy expenditure plus the thermic effect of food.26 Thermic effect of food is relatively small but in some individuals, physical activity energy expenditure may be very significant. Very active individuals generate more metabolic waste than sedentary individuals and by this reasoning may require more dialysis. With increasing catabolism, relating, say to underlying sepsis or extra-renal comorbidity, the requirement for dialysis would also be expected to increase. This can be conjectured as a requirement to normalize dialysis dose by values of Wb in which the value of b is decreasing from b ¼ 1, when the required dialysis dose ¼ Kt/W which is proportional to Kt/V, to b ¼ 0, when the required dialysis dose ¼ Kt (Figure 6). As catabolic rate increases (and b decreases), the disparities relating to gender and body size are amplified. Catabolic patients require more dialysis, and smaller patients and women may require a greater proportional increase. In the general population, basal metabolic rate appears to be around 5% lower in women than in men, at least in sedentary women over 50 years of age.27 This may offset some of the effects we have described. However, in dialysis patients, in whom REE may be increased, a sex difference in basal metabolic rate was not apparent, though the numbers studied were small.28 In addition, recent work in the HD population has suggested that women have an REE (normalized to fat-free mass) that is 6% higher than that in men.14 Women in the high-target group (two-pool Kt/V 1.53) in HEMO study had a 19% lower mortality than those in the standard-target group (two-pool Kt/V 1.16)8 in contrast to the findings in the group overall. Extrapolating our findings concerning equivalent Kt/BSA, a two-pool Kt/V of 1.16 in women is equivalent to approximately 0.96–1.02 in a large male patient, while a two-pool Kt/V of 1.53 is equivalent to approximately 1.26–1.32 in a large male patient. In terms of equivalent Kt/BSA therefore, women in the high-target group received a similar dose to men in the standard group, whereas women in the low-target group could be regarded as having been underdialyzed. Current UK guidelines recommend a minimum equilibrated Kt/V (eKt/V) of 1.2, and suggest that to achieve this 353

original article

EM Spalding et al.: Kt/V underestimates hemodialysis dose

consistently, target eKt/V should be 1.4. If in large men, minimum eKt/V is 1.2 and target eKt/V is 1.4, then our findings suggest that for women, these values should be around 16% higher at 1.39 and 1.62 respectively. By the same reasoning, in small men, values for these parameters should be at least 6% higher at 1.27 and 1.48. These differing eKt/V doses deliver the same dose expressed as Kt/BSA. These suggestions are in line with the recommendations of the recent K-DOQI guidelines for hemodialysis, which advise giving consideration to increasing minimum dialysis dose targets for small patients and for women.29 Further work is required to develop these concepts and to translate them into rigorous outcome-based adequacy targets suitable for clinical usage. MATERIALS AND METHODS Patients We studied all patients receiving HD in the Lister Renal unit on first April 2005. Patients were treated exclusively using high-flux synthetic membranes, ultra-pure water, and bicarbonate buffer. A high proportion received on-line hemodiafiltration. Target Kt/V (two-pool) was 1.2 per HD session.30 Data gathering The following information was collected for all patients from computer-held records: K Age K Sex K Height (cm) K Target weight (W, kg) 10 K V was derived from the Watson formula

Statistical analysis Values are expressed ±1 standard deviation unless otherwise stated. Methods utilized were analysis of variance with Bonferroni method for post hoc comparisons, Students t-test, and w2 analysis as appropriate. Results were deemed to be significant at a P-value of o0.05. The statistical packages used were SPSS 15 and Systat 9. GraphPad was used for curve fitting. DISCLOSURE

All the authors declared no competing interests. REFERENCES 1. 2. 3. 4. 5. 6.

7.

8.

9.

10.

11.

V ¼ 2:447  0:09516A þ 0:1074H þ 0:3362W litres ðmalesÞ

ð2Þ

¼ 2:097 þ 0:1069H þ 0:2466W litres ðfemalesÞ

ð3Þ

12.

13.

where A ¼ age (years), H ¼ height (cm), W ¼ weight (kg) Serial body surface area (BSA (m2)), derived from the Haycock formula31

K

BSA ðm2 Þ ¼ W 0:5378 H 0:3964 0:024265

ð4Þ

where H ¼ height (cm), W ¼ weight (kg) Target Kt

K

17. 18.

Target Kt/BSA Target Kt/Wb

K

19.

20.

where b ¼ 0.67 and other values as appropriate. Estimates of ‘required’ target Kt Estimates of ‘required’ target Kt were calculated using two previously published empirically derived formulae based on BSA.20 These were a double-reciprocal estimate (KtDR) and a multiplicative estimate (KtM). Both are shown below. KtDR ¼ 1=ða þ b=BSAÞ; where a ¼ 0:0069 and b ¼ 0:0237 KtM ¼ a  BSAb ; where a ¼ 34:6 and b ¼ 0:63 354

15. 16.

Kt ¼ 1:2V K

14.

21.

22. 23.

ð5Þ 24.

ð6Þ

Gotch FA, Sargent JA. A mechanistic analysis of the National Cooperative Dialysis Study (NCDS). Kidney Int 1985; 28: 526–534. Daugirdas JT. Simplified equations for monitoring Kt/V, PCRn, eKt/V, and ePCRn. Adv Ren Replace Ther 1995; 2: 295–304. Smye SW, Tattersall JE, Will EJ. Modeling the postdialysis rebound: the reconciliation of current formulas. ASAIO J 1999; 45: 562–567. Gotch FA. Application of urea kinetic modeling to adequacy of CAPD therapy. Adv Perit Dial 1990; 6: 178–180. Wolfe RA, Ashby VB, Daugirdas JT et al. Body size, dose of hemodialysis, and mortality. Am J Kidney Dis 2000; 35: 80–88. Port FK, Ashby VB, Dhingra RK et al. Dialysis dose and body mass index are strongly associated with survival in hemodialysis patients. J Am Soc Nephrol 2002; 13: 1061–1066. Lowrie EG, Chertow GM, Lew NL et al. The urea [clearance  dialysis time] product (Kt) as an outcome-based measure of hemodialysis dose. Kidney Int 1999; 56: 729–737. Eknoyan G, Beck GJ, Cheung AK et al. Effect of dialysis dose and membrane flux in maintenance hemodialysis. N Engl J Med 2002; 347: 2010–2019. Depner T, Daugirdas J, Greene T et al. Dialysis dose and the effect of gender and body size on outcome in the HEMO Study. Kidney Int 2004; 65: 1386–1394. 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. Ravussin E, Lillioja S, Anderson TE et al. Determinants of 24-h energy expenditure in man. Methods and results using a respiratory chamber. J Clin Invest 1986; 78: 1568–1578. Lukaski HC, Johnson PE, Bolonchuk WW et al. Assessment of fat-free mass using bioelectrical impedance measurements of the human body. Am J Clin Nutr 1985; 41: 810–817. Wang Z, Heshka S, Wang J et al. Metabolically active portion of fat-free mass: a cellular body composition level modeling analysis. Am J Physiol Endocrinol Metab 2007; 292: E49–E53. Sarkar SR, Kuhlmann MK, Kotanko P et al. Metabolic consequences of body size and body composition in hemodialysis patients. Kidney Int 2006; 70: 1832–1839. Heusner AA. Body size and energy metabolism. Annu Rev Nutr 1985; 5: 267–293. Singer MA, Morton AR. Mouse to elephant: biological scaling and Kt/V. Am J Kidney Dis 2000; 35: 306–309. White CR, Seymour RS. Allometric scaling of mammalian metabolism. J Exp Biol 2005; 208: 1611–1619. Cunningham JJ. Body composition and resting metabolic rate: the myth of feminine metabolism. Am J Clin Nutr 1982; 36: 721–726. Lowrie EG, Li Z, Ofsthun N et al. Measurement of dialyzer clearance, dialysis time, and body size: death risk relationships among patients. Kidney Int 2004; 66: 2077–2084. Lowrie EG, Li Z, Ofsthun N et al. The online measurement of hemodialysis dose (Kt): clinical outcome as a function of body surface area. Kidney Int 2005; 68: 1344–1354. Kotanko P, Thijssen S, Kitzler T et al. Size matters: body composition and outcomes in maintenance hemodialysis patients. Blood Purif 2007; 25: 27–30. Suri RS, Garg AX, Chertow GM et al. Frequent Hemodialysis Network (FHN) randomized trials: study design. Kidney Int 2007; 71: 349–359. Leavey SF, McCullough K, Hecking E et al. Body mass index and mortality in ‘healthier’ as compared with ‘sicker’ haemodialysis patients: results from the Dialysis Outcomes and Practice Patterns Study (DOPPS). Nephrol Dial Transplant 2001; 16: 2386–2394. Kalantar-Zadeh K, Kopple JD, Kilpatrick RD et al. Association of morbid obesity and weight change over time with cardiovascular

Kidney International (2008) 74, 348–355

original article

EM Spalding et al.: Kt/V underestimates hemodialysis dose

25.

26.

27.

survival in hemodialysis population. Am J Kidney Dis 2005; 46: 489–500. Lowrie EG, Li Z, Ofsthun NJ et al. Evaluating a new method to judge dialysis treatment using online measurements of ionic clearance. Kidney Int 2006; 70: 211–217. Toth MJ. Comparing energy expenditure data among individuals differing in body size and composition: statistical and physiological considerations. Curr Opin Clin Nutr Metab Care 2001; 4: 391–397. Ferraro R, Lillioja S, Fontvieille AM et al. Lower sedentary metabolic rate in women compared with men. J Clin Invest 1992; 90: 780–784.

Kidney International (2008) 74, 348–355

28. 29. 30.

31.

Ikizler TA, Wingard RL, Sun M et al. Increased energy expenditure in hemodialysis patients. J Am Soc Nephrol 1996; 7: 2646–2653. Clinical practice guidelines for hemodialysis adequacy, update 2006. Am J Kidney Dis 2006; 48(Suppl 1): S2–S90. Tattersall JE, DeTakats D, Chamney P et al. The post-hemodialysis rebound: predicting and quantifying its effect on Kt/V. Kidney Int 1996; 50: 2094–2102. Haycock GB, Schwartz GJ, Wisotsky DH. Geometric method for measuring body surface area: a height-weight formula validated in infants, children, and adults. J Pediatr 1978; 93: 62–66.

355