Scaling Hemodialysis Dose: Kt Over What?

Abstract

Related Article, p. 358 Related Article, p. 358 Dialysis dose, like kidney function, is commonly expressed as a clearance scaled to some indicator of body size. With glomerular filtration rate (GFR), the traditional scaling parameter is body surface area (BSA), whereas dialysis dose, expressed as fractional clearance of urea, usually is scaled to total-body water volume, which is similar in size to the urea distribution volume, V. Results from the Hemodialysis (HEMO) Study, in which there was a possible benefit associated with a higher target Kt/V in women but not men, suggested that the concept of scaling Kt to V may not optimally account for important patient characteristics that determine the necessary dialysis dose.1Depner T. Daugirdas J. Greene T. et al.Hemodialysis Study GroupDialysis dose and the effect of gender and body size on outcome in the HEMO Study.Kidney Int. 2004; 65: 1386-1394Abstract Full Text Full Text PDF PubMed Scopus (153) Google Scholar In this issue of the AJKD, Sridharan et al,2Sridharan S. Vilar E. Davenport A. et al.Scaling hemodialysis target dose to reflect body surface area, metabolic activity, and protein catabolic rate: a prospective, cross-sectional study.Am J Kidney Dis. 2017; 69: 358-366Google Scholar motivated by the results in HEMO and several observational studies showing worse outcomes in smaller hemodialysis patients, examine scaling of dialysis dose in 1,500 hemodialysis patients and consider the effects of alternative scaling denominators. In this patient cohort, Sridharan et al calculated a patient-specific Kt value that would correspond to a single-pool Kt/V of 1.2. This Kt value then was rescaled to a variety of sizing parameters, including BSA (Haycock equation), resting energy expenditure (REE, calculated from an equation derived in patients with chronic kidney disease), total energy expenditure (TEE, equal to REE multiplied by a metabolic activity ratio derived from a questionnaire about daily physical activity), and normalized protein catabolic rate (nPCR). They concluded that use of most of these alternative scaling measures would result in the need for a relatively higher dialysis dose (as Kt) in certain groups of patients, including women and small patients, younger patients, patients with low comorbidity, employed patients (due to their higher activity level), and patients of South Asian ethnicity. Adjustments that would need to be made in women, patients with lower body weight, and younger versus older patients are shown in Table 1.Table 1Rescaling of Dialysis Dose to BSA, REE or TEE, or nPCRKt/VKtKt/BSAKt/REEKt/TEEKt/nPCRMale sex1.249.325.430.325.651.1Female sex1.238.721.927.223.440.2Delta, %021.513.810.28.5921.3Weight > 85 kg1.255.625.130.125.557.9Weight < 62 kg1.235.622.828.124.136.8Delta, %0.036.09.166.743.9836.4Age < 52 y1.246.924.728.123.046.7Age > 75.5 y1.242.223.430.326.445.6Delta, %010.05.26−7.83−14.82.36Note: Weight and age show highest versus lowest quartile. Delta % is calculated relative to the top row in each category. Data are calculated from selected mean values in Table 2 in Sridharan et al.2Sridharan S. Vilar E. Davenport A. et al.Scaling hemodialysis target dose to reflect body surface area, metabolic activity, and protein catabolic rate: a prospective, cross-sectional study.Am J Kidney Dis. 2017; 69: 358-366Google ScholarAbbreviations: BSA, body surface area; nPCR, normalized protein catabolic rate; REE, resting energy expenditure; TEE, total energy expenditure. Open table in a new tab Note: Weight and age show highest versus lowest quartile. Delta % is calculated relative to the top row in each category. Data are calculated from selected mean values in Table 2 in Sridharan et al.2Sridharan S. Vilar E. Davenport A. et al.Scaling hemodialysis target dose to reflect body surface area, metabolic activity, and protein catabolic rate: a prospective, cross-sectional study.Am J Kidney Dis. 2017; 69: 358-366Google Scholar Abbreviations: BSA, body surface area; nPCR, normalized protein catabolic rate; REE, resting energy expenditure; TEE, total energy expenditure. How do we interpret the numbers in Table 1? Focusing on sex differences, the results suggest that if dialysis were rescaled to BSA, at equivalent levels of Kt/V, women would receive a 14% lower dose of dialysis than men. If dialysis dose were rescaled to REE or TEE, women would receive a 10% or 9% lower dose, respectively, and if dialysis dose were rescaled to nPCR, women would receive a 21% lower dose. Rescaling of dialysis to BSA, REE, TEE, or nPCR would also result in a lower dose to lighter patients: 9% with rescaling to BSA, 6.7% with rescaling to REE, 4% with rescaling to TEE, and 36% with rescaling to nPCR. With regard to age, comparing patients younger than 52 years versus those older than 75 years, rescaling to BSA or nPCR would slightly (5.3% or less) increase the effective dose given to younger patients, whereas rescaling based on energy expenditure would result in an 8% lower dose in the younger group based on REE and a 15% lower dose in the younger patients based on TEE. When examining the rescaling effects shown in Table 1, the needed dose adjustments in the various “delta” rows are small, all <15% (with the exception of Kt/nPCR); a ≤15% difference in dialysis dose may seem to be of insufficient magnitude to be clinically important. However, one could argue that the dose scaling corrections determined in these analyses should be applied not to the treatment Kt/V, but to a continuous equivalent measure of dialysis dose such as standard Kt/V. With the usual 3-sessions-per-week clinical schedule, in order to effect a 15% change in standard Kt/V, one needs to change session Kt/V by twice as much; that is, by ∼30%. For example, in the HEMO Study, separation between the high- and low-dose groups was 30% based on equilibrated session Kt/V, but only 17% based on standard Kt/V.3Daugirdas J.T. Greene T. Chertow G.M. Depner T.A. Can rescaling dose of dialysis to body surface area in the HEMO Study explain the different responses to dose in women versus men?.Clin J Am Soc Nephrol. 2010; 5: 1628-1636Crossref PubMed Scopus (59) Google Scholar As an example, the 14% “underdialysis” in women calculated in Table 1 based on BSA rescaling, if thought of as a 14% shortfall in standard Kt/V, would require a 28% higher dose of session Kt/V to correct. In other words, if the current spKt/V target in men is given as 1.3, in women, it should be 1.3 × 128% = 1.7. To what extent are the hypothesized rescaling strategies advocated by Sridharan et al supported by what we know from physiologic studies? Dialysis is “kidney replacement therapy,” so it is not unreasonable to scale hemodialysis to body size in ways that are similar to scaling of normal kidney function. Scaling dialysis clearance (Kt) to BSA is consistent with normal physiology because both measured GFR and estimated GFR normally scale to BSA. When GFR is measured using radiolabeled iothalamate, mean GFR scaled to BSA is very similar in men and women, whereas iothalamate-measured GFR scaled to anthropometric estimates of total-body water is markedly different between the sexes.4Daugirdas J.T. Meyer K. Greene T. Butler R.S. Poggio E.D. Scaling of measured glomerular filtration rate in kidney donor candidates by anthropometric estimates of body surface area, body water, metabolic rate, or liver size.Clin J Am Soc Nephrol. 2009; 4: 1575-1583Crossref PubMed Scopus (27) Google Scholar If one considers GFR from infancy through childhood to adulthood, GFR/BSA is close to 100 mL/min/1.73 m2 beginning at approximately age 2 years.5Daugirdas J.T. Hanna M.G. Becker-Cohen R. Langman C.B. Dose of dialysis based on body surface area is markedly less in younger children than in older adolescents.Clin J Am Soc Nephrol. 2010; 5: 821-827Crossref PubMed Scopus (25) Google Scholar It is attractive to link uremic toxin generation to REE because the portion of uremic toxin generation that does not come from toxins absorbed from the gut must come from metabolic activity in the patient’s own cells. Equations that estimate REE, such as the Vilar equation6Vilar E. Machado A. Garrett A. Kozarski R. Wellsted D. Farrington K. Disease-specific predictive formulas for energy expenditure in the dialysis population.J Ren Nutr. 2014; 24: 243-251Abstract Full Text Full Text PDF PubMed Scopus (28) Google Scholar used by Sridharan et al, tend to have both an age component and a sex component. If one looks at the ratio of iothalamate-estimated GFR to REE in healthy kidney donors,4Daugirdas J.T. Meyer K. Greene T. Butler R.S. Poggio E.D. Scaling of measured glomerular filtration rate in kidney donor candidates by anthropometric estimates of body surface area, body water, metabolic rate, or liver size.Clin J Am Soc Nephrol. 2009; 4: 1575-1583Crossref PubMed Scopus (27) Google Scholar the ratio is not sex neutral, though the disparity is less than the difference in sex ratio of measured GFR to V, where V is determined by the Watson formula.4Daugirdas J.T. Meyer K. Greene T. Butler R.S. Poggio E.D. Scaling of measured glomerular filtration rate in kidney donor candidates by anthropometric estimates of body surface area, body water, metabolic rate, or liver size.Clin J Am Soc Nephrol. 2009; 4: 1575-1583Crossref PubMed Scopus (27) Google Scholar As shown in Table 1, rescaling of dialysis dose to BSA would require a small increase in dialysis dose for the elderly (5.3% at equivalent levels of Kt/V) because BSA apparently did not decrease with age as much as does total-body water (V). However, REE rescaling would justify a 7.8% reduction in target dose in the elderly because the decrease in REE with age is steeper than the decrease in V.4Daugirdas J.T. Meyer K. Greene T. Butler R.S. Poggio E.D. Scaling of measured glomerular filtration rate in kidney donor candidates by anthropometric estimates of body surface area, body water, metabolic rate, or liver size.Clin J Am Soc Nephrol. 2009; 4: 1575-1583Crossref PubMed Scopus (27) Google Scholar Sridharan et al take metabolic rate scaling beyond REE and hypothesize that dialysis dose might need to be scaled not to REE, but rather to TEE. They estimated TEE in their patients using an assessment of physical activity that was used to compute a multiplier for REE. Although there is at least one observational study suggesting that kidney function (as estimated GFR7Hawkins M.S. Sevick M.A. Richardson C.R. Fried L.F. Arena V.C. Kriska A.M. Association between physical activity and kidney function: National Health and Nutrition Examination Survey.Med Sci Sports Exerc. 2011; 43: 1457-1464Crossref PubMed Scopus (82) Google Scholar) is 3% to 6% higher in people who are more physically active, I know of no studies in healthy patients comparing GFR while sedentary versus while undergoing moderate physical activity. Also, in general populations, estimated or measured GFR is not adjusted based on current physical activity level. In the analysis by Sridharan et al, the effect of normalizing Kt to TEE in terms of sex and body size was qualitatively similar to normalizing to REE for most subgroups. Scaling to TEE could justify a reduction in dialysis dose for the elderly beyond scaling to REE, due to the relative lack of physical activity in the older age group. The inference that more dialysis may be needed for patients who are employed or with low comorbidity is based primarily on the hypothesis that patients with higher TEE may require more dialysis, which in my opinion is not supported by physiologic analysis, although it is possible that persons with a high TEE may eat more protein and other foods, perhaps resulting in a need for increased dialysis dose by this mechanism. The scaling factor used by Sridharan et al for protein intake, nPCR, is somewhat problematic. nPCR is calculated as 5.42 × (G/V) + 0.17, where G is urea generation rate.8Depner T.A. Approach to hemodialysis kinetic modeling.in: Henrich W.L. Principles and Practice of Dialysis. 4th ed. Lippincott Williams & Wilkins, Philadelphia, PA2009: 73Google Scholar G is closely related to the unscaled PCR because the major source of nitrogen excreted as urea comes from protein. G itself tends to scale to V, and because the authors are using a scaling factor of G/V when they scale to nPCR, they are scaling to a variable that is related to V and are then dividing by V. As a result, scaling Kt to nPCR results in loss of body size adjustment. This strategy tends to give a one-size-fits-all dose of dialysis (Kt) for patients regardless of size. When nPCR is small, that is, less than 1.0, there should be less of a metabolic challenge in terms of nitrogenous waste products and physiologically, the dose of dialysis (as Kt) perhaps should be reduced. In contrast, when the authors rescale to Kt divided by nPCR, low nPCR requires a higher dose of dialysis. An argument could be made that a low nPCR necessitates more dialysis because underdialysis could be contributing to a low nPCR by causing uremic anorexia; however, increasing Kt in the setting of low nPCR would reflect a therapeutic decision rather than a physiologic—that is, nitrogenous uremic toxin-based–rationale for a higher dialysis dose. In summary, the analyses and data of Sridharan et al are an interesting thought exercise that questions whether the current approach of scaling of dialysis dose (as Kureat) to total-body water is the correct strategy. Given the difficulties interpreting observational data in hemodialysis due to dose-targeting bias,9Greene T. Daugirdas J. Depner T. et al.Hemodialysis Study GroupAssociation of achieved dialysis dose with mortality in the Hemodialysis Study: an example of “dose-targeting bias.”.J Am Soc Nephrol. 2005; 16: 3371-3380Crossref PubMed Scopus (74) Google Scholar it is unlikely that preferential use of any one of these alternative scaling strategies will be supported by high-level evidence (eg, primary outcome of randomized trials). Thus, the choice of what to do should remain with the nephrology caregiver. Nevertheless, because both BSA and energy expenditure rescaling support giving more dialysis to smaller patients and to women and there are some clinical outcomes data suggesting that women may require more dialysis than men in terms of Kt/V,1Depner T. Daugirdas J. Greene T. et al.Hemodialysis Study GroupDialysis dose and the effect of gender and body size on outcome in the HEMO Study.Kidney Int. 2004; 65: 1386-1394Abstract Full Text Full Text PDF PubMed Scopus (153) Google Scholar, 3Daugirdas J.T. Greene T. Chertow G.M. Depner T.A. Can rescaling dose of dialysis to body surface area in the HEMO Study explain the different responses to dose in women versus men?.Clin J Am Soc Nephrol. 2010; 5: 1628-1636Crossref PubMed Scopus (59) Google Scholar perhaps we should view dialysis adequacy in women and in smaller patients with concern when their current prescription provides only a minimally adequate level of Kt/V. Support: None. Financial Disclosure: The author declares that he has no relevant financial interests. Peer Review: Evaluted by Deputy Editor Weiner and Editor-in-Chief Levey. Scaling Hemodialysis Target Dose to Reflect Body Surface Area, Metabolic Activity, and Protein Catabolic Rate: A Prospective, Cross-sectional StudyAmerican Journal of Kidney DiseasesVol. 69Issue 3PreviewWomen and small men treated by hemodialysis (HD) have reduced survival. This may be due to use of total-body water (V) as the normalizing factor for dialysis dosing. In this study, we explored the equivalent dialysis dose that would be delivered using alternative scaling parameters matching the current recommended minimum Kt/V target of 1.2. Full-Text PDF