Transcript
  • SPECIAL ARTICLE

    Blood Pressure Control in Dialysis Patients:Importance of the Lag Phenomenon

    Bernard Charra, MD, Jonas Bergstrom, MD, and Belding H. Scribner, MD

    c Failure by the world dialysis community to understand and use the dry-weight method of blood pressure (BP)control has resulted in an increasing incidence of treatment-resistant hypertension, which remains the principalcause of cardiovascular morbidity and mortality. This failure may in part be because the relationship between theextracellular volume (ECV) and BP is not simple and linear, but complex, because of a lag of several weeks betweenthe normalization of the time-averaged ECV and the decrease in BP. Another cause for this failure may be theunwillingness to taper and stop all antihypertensive medications during the transition from hypertension tonormotension. In this report, we describe in detail the lag phenomenon, document its presence during treatment inother populations, and describe how this knowledge is used in the application of the dry-weight method of drug-freeBP control in the dialysis population.r 1998 by the National Kidney Foundation, Inc.

    INDEX WORDS: Hemodialysis; hypertension; dry weight; lag time; extracellular volume.

    ANUMBER OF RECENT publications showthat the procedure for controlling bloodpressure (BP) using the dry-weight method1 isnot well understood. The purpose of this report isto show that with the dry-weight method of BPcontrol, the relationship between BP and extracel-lular volume (ECV) is not linear, but complex,because of what we term the lag phenomenon.

    BRIEF HISTORY

    In 1960, one of the authors (B.H.S.) and hiscolleagues saved the life of the first chronicdialysis patient by curing his malignant hyperten-sion using aggressive ultrafiltration.2

    In the 1970s, long dialysis sessions combinedwith a low-salt diet allowed the control of hyper-tension without antihypertensive medication in90% of the patients.3-5 In recent years, shorteningof dialysis time has led to a resurgence of hyper-tension.6,7 Today, hypertension has become a

    leading cause of morbidity and mortality in thedialysis population.8,9

    The entire Tassin experience with BP control,which began more than 30 years ago, is shown inFig 1. It involves 712 patients. A detailed descrip-tion of these patients has been published else-where.10 Each patients record showed this lagphenomenon during what we term the probe fordry weight. However, until now, we and otherinvestigators who have encountered this lag phe-nomenon11,12 have failed to delineate it or recog-nize its importance in understanding and imple-menting the dry-weight method of BP control,especially in the dialysis population.

    DESCRIPTION OF THE LAG PHENOMENON

    Figure 1 shows the overall experience in Tas-sin with drug-free BP control using the dry-weight method in 712 patients.10 Note that themean loss of predialysis weight from 64.3 kg(95% confidence interval [CI], 63.7 to 64.8 kg)to 62 kg (95% CI, 61.4 to 62.6 kg) is alreadycomplete by the end of the first month (P 50.005), indicating, we believe, a decrease ofabout 2 L in the predialysis ECV. A recent studyhas confirmed, by direct measurement of theECV, an approximately 2- to 3-L difference be-tween hypertensive and normotensive dialysispatients.13

    During this same month, the mean arterialpressure (MAP) only decreased from 121 mmHg (95% CI, 119.0 to 122.6 mm Hg) to 108 mmHg (95% CI, 106.6 to 109.3 mm Hg). At this

    From the Centre de rein artificiel, Tassin, France; Depart-ment of Clinical Sciences, Division of Baxter Novum, Karo-linska Institutet, Huddinge, Sweden; and the University ofWashington School of Medicine, Seattle, WA.

    Received December 18, 1997; accepted in revised formApril 24, 1998.

    Presented in part at the American Society of Nephrology30th Annual Meeting, San Antonio, TX, November 2-5,1997.

    Address reprint requests to Bernard Charra, MD, Centrede rein artificiel, 42 Avenue du 8-Mai-1945, 69160 Tassin,France. E-mail: [email protected]

    r 1998 by the National Kidney Foundation, Inc.0272-6386/98/3205-0003$3.00/0

    720 American Journal of Kidney Diseases, Vol 32, No 5 (November), 1998: pp 720-724

  • point, antihypertensive medications were still be-ing withdrawn.

    The MAP continued to decrease for another 8months despite the withdrawal of antihyperten-sive medication in more than 90% of the 712patients. It was this delay between the abruptnormalization of ECV and the much more gradualdecrease in the predialysis MAP that we havetermed the lag phenomenon.

    After the first month, serial changes in postdi-alysis weight no longer reflect the changes inECV because the patient becomes anabolic andgains dry weight. This increase in dry weight isconsistent with a gradual increase in predialysisserum creatinine level from 830 to 930 mol/Lover the next 12 months (P , 0.001), despite aconstant dose of dialysis,10 which maintains theurea Kt/V at greater than 1.6.

    Figure 2 shows in more detail what happenedto ECV as reflected in body weight and MAPduring the first month in 96 patients started ondialysis in Tassin since 1992. In this figure, weused predialysis weight to reflect the size of theECV before treatment began. Figure 2 shows theweight/ECV decreasing by approximately 2 Learly in the first month. Meanwhile, MAP de-creased gradually over the same period, duringwhich time antihypertensive medications weretapered.

    Figure 3 shows a free graph of this lag phenom-enon as we envision it. The predialysis ECVdecreased promptly, reaching a low within the

    first month, after which it remained stable. Thepredialysis MAP took months to finally stabilize,long after antihypertensive medications had beenwithdrawn.

    OTHER EXAMPLES OF THE LAGPHENOMENON

    It is relevant that in the 1940s, Kempner11 wasable to treat essential hypertension successfullyby dietary sodium restriction alone using hisrice-fruit diet, which was virtually sodium free.His success rate exceeded 50%, despite the al-most impossible rigors of this difficult diet. Thelag between the institution of this diet and thedecrease in BP was at least 1 month. Directmeasurement by Murphy14 of the ECV beforeand after the institution of the rice-fruit diet in 17patients showed a very similar decrease to whatwe have seen of approximately 2 L.

    Another example of this lag phenomenon oc-curs when diuretics alone are used to treat hyper-tension. Responders have an initial ECV de-crease of 1.5 to 2 L. The BP begins to decrease

    Fig 1. Postdialysis average weight (kilograms) andpredialysis average MAP (millimeters of mercury) in712 Tassin patients in the 12 first dialysis months. Theline represents predialysis MAP 1 1 standard error ofthe mean (SEM); the bars represent the mean postdialy-sis weight 1 1 SEM; (O), postdialysis weight; (r),predialysis MAP. Abbreviation: AntiHT, antihyperten-sive.

    Fig 2. Predialysis average weight (kilograms) andpredialysis average MAP (millimeters of mercury) in 96Tassin patients in the first hemodialysis month. Theline represents predialysis MAP 1 1 SEM; the barsrepresent the mean postdialysis weight 1 1 SEM; (O),predialysis weight; the bars represent predialysis MAP.

    Fig 3. Theoretical graph of predialysis ECV andpredialysis MAP during the first 12 months of hemodi-alysis. The grey line represents predialysis ECV; theblack line, the predialysis MAP.

    LAG PHENOMENON 721

  • within the first week and continues to decreaseover 2 to 3 months.12

    THE ECV AT DRY WEIGHT

    At dry weight, the ECV as measured by stan-dard isotopic methods is in the normal range.15However, we believe that patients at dry weighthave an ECV that is at the low end of the normalrange. In this respect, dialysis patients may besimilar to such populations as the YanomamoIndians, who ingest an almost sodium-free dietand have no hypertension.16 At dry weight, apatient is completely intolerant to antihyperten-sive medications.

    Serial measurements of the ECV during thelag period using newer methodology, such asbioimpedance,17 may help clarify this dynamicrelationship between ECV and BP. The clinicalavailability of such measurements may make iteasier to achieve dry weight during the probe.

    PROBING FOR DRY WEIGHT

    From the practical standpoint, instituting thedry-weight method of drug-free BP control in adialysis patient is not easy, especially during thephase of normalizing BP while withdrawing BPmedications. We call this phase probing for dryweight.1

    What typically happens during the first monthis shown in Fig 2. Before the start of dialysistreatment, 90% of our patients are hypertensive,despite receiving antihypertensive medication.Their ECV is expanded, even though most haveno edema.

    At the start of the probe for dry weight, dialy-sis sessions are increased from 3 to 8 hours inlength in 1-hour increments. Intense, carefullymonitored ultrafiltration plus a strict low-sodiumdiet permit a gradual reduction in the predialysisweight of approximately 2 kg and postdialysisweight of 3 kg over the first 2 to 4 weeks. Theactual rate of decrease is strictly by trial anderror, governed by the patients tolerance, toreduce to a minimum episodes of muscle crampsand hypotension.

    During this initial period, antihypertensivemedications are gradually withdrawn. Failure todo so makes it impossible to achieve and main-tain dry weight.

    During this first month, the patients appetitegradually improves. They become anabolic and

    begin to put on real body weight. This changesoon begins to complicate the problem of deter-mining postdialysis weight, which calls for care-ful medical judgment and supervision.

    Obviously, the probe for dry weight representsa difficult transition for the patients. This transi-tion must be carefully explained to them andongoing support from physicians and staff areabsolutely essential. At the same time, the pa-tients must be made to understand that once thistransition is over, they will feel much better. Thehighly restrictive low-sodium diet can be liberal-ized somewhat.

    It is important to explain to the patient that ifthe initial rigidly restricted low-sodium diet isfollowed, the blandness of food will disappearwithin some weeks. The pleasant taste of un-salted food will be rediscovered and patientsoften complain about the burning sensation whennormally salted food is ingested. We believe thisresetting of the taste for salt explains the veryrare complaint of thirst by our patients, as well asthe mean observed interdialytic weight gain ofonly 1.6 kg.

    The patient also must understand that, afterreaching dry weight, he no longer will have totake antihypertensive medications, which, in thedialysis patient, are poorly effective and, in thedosages used, often have debilitating side effects.Most important, it must be explained that bycuring the hypertension, the danger of having aheart attack or a stroke is greatly reduced.

    PATHOPHYSIOLOGY OF HYPERTENSIONIN THE DIALYSIS PATIENT

    We postulate that during the probe for dryweight, the reverse sequence of hemodynamicmodifications after a saline load in patients withend-stage renal failure is experienced, as de-scribed by Guyton.18 This investigator has shownthat a sodium load results in a sequence thatincludes ECV overload, transient increased car-diac output, and increased total peripheral resis-tance resulting in increased BP. Hypertension, inturn, increases the natriuresis that returns theECV to an almost normal level. The difference inECV between normotension and hypertensiondoes not exceed 2% of body weight.18

    722 CHARRA, BERGSTRO M, AND SCRIBNER

  • DISCUSSION

    Failure to understand that relatively smallchanges in time-averaged ECV, if sustained, canhave profound effects on BP because of gradualchanges in peripheral resistance has led to ques-tions about the validity of the dry-weight meth-od.19 However, this conclusion was mainly basedon acute studies that showed poor correlationbetween increase in BP and weight gain duringone interdialytic period.20,21 Hence, these studiesdid not consider the lag phenomenon. Indeed,other studies22 using noninvasive technologies ofvolume assessment concluded that ECV over-load has an essential role in dialysis hyperten-sion.

    Even more to the point, we believe that arecently completed report on BP control in thedialysis patient by a prestigious committee23should have acknowledged the scientific validityof the dry-weight method of BP control as prac-ticed in Tassin. One does not need a double-blindstudy to prove that the dry-weight method isscientifically valid any more than Kempner11needed a double-blind study to prove that therice-fruit diet could cure essential hypertension,because in those days, there was no other cure forhypertension.

    The same dilemma must be faced by the taskforce of the National Kidney Foundation, whichcurrently is finalizing a report on how to stem theepidemic of cardiovascular complications in theend-stage renal disease population. At the basiclevel, the answer is quite simple. Control of BPusing the dry-weight method will markedly re-duce the incidence of cardiovascular complica-tions, as the Tassin experience clearly shows.10

    As we see it, the problem is how to adaptdrug-free, dry-weight control of BP to todaysincenter dialysis practice. Recent publications24-26have shown that skillful manipulation of dialy-sate sodium concentration improves BP controlin patients undergoing short incenter dialysis.More attention to lowering dietary sodium intakealso improves BP control.26

    The recent use of long, very slow home dialy-sis sessions by Pierratos et al27 represents an-other solution to this problem. More extensiveuse of chronic ambulatory peritoneal dialysis inthe early stages of end-stage renal disease dia-lytic therapy should provide an excellent way toimplement the dry-weight method of drug-free

    BP control.28 Other solutions may be found oncethere is again proper funding for clinical investi-gation in this area.2

    In the meantime, a good beginning might be tostudy and refine old and new dietary methods forreducing the sodium intake of the US dialysispopulation. For every gram that a patients so-dium intake is reduced between each dialysis, theneed to remove ECV by ultrafiltration also isreduced by approximately 300 mL. Also, thereduction in cramps and hypotension during shortdialysis26 would serve as an incentive to patientsto comply with more sodium restriction in thediet.

    REFERENCES1. Charra B, Chazot C, Laurent G, Calemard E, Terrat JC,

    Vanel T, Jean G, Ruffet M: Clinical assessment of dryweight. Nephrol Dial Transplant 11:16-19, 1996 (suppl 2)

    2. Scribner BH: Milestones in nephrology: The treatmentof chronic uremia by means of intermittent hemodialysis. Apreliminary report. Trans Am Soc Artif Intern Organs 6:114-122, 1960

    3. Hegstrom RM, Murray JS, Pendras JP, Burnell JM,Scribner BH: Hemodialysis in the treatment of chronicuremia. Trans Am Soc Artif Int Organs 7:136-149, 1961

    4. Comty C, Rottka H, Shaldon S: Blood pressure controlin patients with end-stage renal failure treated by intermit-tent dialysis. Proc Eur Dial Transplant Assoc 1:209-210,1964

    5. Vertes V, Cangiano JL, Berman LB, Gould A: Hyper-tension in end-stage renal disease. N Engl J Med 280:978-981, 1969

    6. Wizemann V, Kramer W: Short-term dialysis, long-term complications. Blood Purif 5:193-201, 1987

    7. Cheigh JS, Milite C, Sullivan JF, Rubin AL, StenzelKH: Hypertension is not adequately controlled in hemodialy-sis patients. Am J Kidney Dis 19:453-459, 1992

    8. Salem MM, Bower JD: Hypertension in the hemodialy-sis population. Am J Kidney Dis 29:811-812, 1997

    9. Foley RN, Parfrey PS, Harnett JD, Kent GM, MurrayDC, Barre PE: Impact of hypertension on cardiomyopathy,morbidity and mortality in end-stage renal disease. KidneyInt 49:1379-1385, 1996

    10. Charra B, Calemard E, Laurent G: Importance oftreatment time and blood pressure control in achievinglong-term survival on dialysis. Am J Nephrol 16:35-44,1996

    11. Kempner W: Treatment of hypertensive vascular dis-ease with rice diet. Am J Med 8:545-577, 1945

    12. Freis ED, Reda DJ, Materson BJ: Volume (weightloss) and blood pressure response following thiazide diuret-ics. Hypertension 12:244-250, 1988

    13. Katzarski KS, Charra B, Luik A, Habets J, Nisell J,Filho JD, Leypoldt JK, Leunissen KML, Laurent G, Berg-strom J: Fluid state, blood volume and blood pressurecontrol in patients treated with long hemodialysis procedure.

    LAG PHENOMENON 723

  • A comparison with conventional hemodialysis treatment.Nephrol Dial Transplant (in press)

    14. Murphy RJF: The effect of rice diet on plasmavolume and extracellular fluid space in hypertensive sub-jects. J Clin Invest 29:912-917, 1950

    15. Blumberg A, Nelp WD, Hegstrom RM, Scribner BH:Extracellular volume in patients with chronic renal diseasetreated for hypertension by sodium restriction. Lancet 2:69-73, 1967

    16. Freis ED: Salt, volume and the prevention of hyper-tension. Circulation 53:589-595, 1976

    17. de Vries PMJM: Plasma volume changes during he-modialysis. Semin Dial 5:42-47, 1992

    18. Guyton AC: Renal functional curve: A key to under-standing the pathogenesis of hypertension? Hypertension10:1-6, 1987

    19. Kirchner KA: Hypertension in hemodialysis patients:More questions than answers. Am J Kidney Dis 30:577-578,1997

    20. Luik AJ, van Kujik WHM, Spek J, de Heer F, vanBortel LMA, Schiffers PMH, van Hoof JP, Leunissen KML:Effects of hypervolemia on interdialytic hemodynamics andblood pressure control in hemodialysis patients. Am J Kid-ney Dis 30:466-474, 1997

    21. Savage T, Fabbian F, Giles M, Tomson CRV, RaineAEG: Interdialytic weight gain and 48-hour blood pressurein haemodialysis patients. Nephrol Dial Transplant 12:2308-2311, 1997

    22. Katzarski KS, Nisell J, Randmaa I, Danielsson A,Freyschuss U, Bergstrom J: A critical evaluation of ultra-sound measurement of inferior vena cava diameter in assess-ing dry weight in normotensive and hypertensive hemodialy-sis patients. Am J Kidney Dis 30:459-465, 1997

    23. Mailloux LU, Haley W: Hypertension in the ESRDpatient: Pathophysiology, therapy, outcomes, and future di-rections. Am J Kidney Dis 32:705-719, 1998

    24. Flanigan MJ, Khairullah QT, Lim VS: Dialysatesodium delivery can alter chronic blood pressure manage-ment. Am J Kidney Dis 29:383-391, 1997

    25. Donohoe P, Farmer C, Dallyn P, Kingswood JC,Goldsmith DJA, Sharpstone P: Low-sodium hemodialysiswithout fluid removal improves blood pressure control inchronic dialysis patients. Kidney Int 52:1119, 1997 (abstr)

    26. Krautzig S, Janssen U, Koch KM, Granoleras C,Shaldon S: Dietary salt restriction and reduction of dialysatesodium to control hypertension in maintenance haemodialy-sis patients. Nephrol Dial Transplant 13:552-553, 1998

    27. Pierratos A, Ouwendyk M, Francoeur R, Vas S, RajD, Ecclestone AM, Langos V, Uldall R: Nocturnal hemodi-alysis: Three-year experience. J Am Soc Nephrol 9:859-868,1998

    28. Young MA, Nolph KD, Dutton S, Prowant BF: Anti-hypertensive drug requirements in CAPD. Perit Dial Bull4:85-88, 1984

    724 CHARRA, BERGSTRO M, AND SCRIBNER


Top Related