Can non-pharmacological interventions reduce doses of drugs needed for the treatment of hypertension? World Hypertension League.

Update/Le point Can non-pharmacological interventions reduce doses of drugs needed for the treatment of hypertension?* World Hypertension League1 This article describes the impact of non-pharmacological interventions-salt restriction, weight control, alcohol consumption, exercise, and potassium supplementation-on the effects of drug treatment of hypertension, and their potential to reduce the patient's drug requirements. Although consensus on all aspects of such interventions has not yet been reached, the evidence that these approaches may reduce the need for drugs is considered satisfactory. *A French translation of this article will appear in a later issue of the Bulletin. Requests for reprints should be sent to the World Hypertension League, 20 Avenue du Bouchet, 1209 Geneva, Switzerland. A slightly different version of this paper has been accepted for publication in the Journal of human hypertension. 1 The World Hypertension League is an association of anti- hypertensive leagues, hypertension societies, committees and other national bodies whose aim is to control hypertension in populations. The present document is based on a background paper prepared by F.C. Luft, University Erlangen-Nurnberg, whose work is gratefully acknowledged. It is one of a series of statements on topics that are of practical importance for the management of hypertension, addressed to practising physi- cians. Since the aim is to present practical information, only selected bibliography is quoted. Like the earlier papers, the text has been repeatedly and extensively discussed by represen- tatives of the League's member organizations and accepted as a consensus document. In May 1992 the World Hypertension League comprised the following member organizations: All India Heart Foundation; Austrian Hypertension Society (Austrian Hypertension League); Belgium Hypertension Committee; Brazilian Society of Hyperten- sion; British Hypertension Society; Canadian Coalition for High Blood Pressure Prevention and Control; Chilean Society of Hypertension; Chinese Hypertension Society; Council for High Blood Pressure of the Irish Heart Foundation; Czech and Slovak Committee of Hypertension; Danish Society of Hypertension; Dutch Hypertension Society; French National Committee against Arterial Hypertension; German League against Hypertension; Hellenic Society of Hypertension; Hypertension League of Morocco; Hypertension Society of the Hungarian Society of Cardiology; Inter-American Society of Hypertension; Israel Hypertension Society; Italian League against Hypertension; Japan Heart Foundation; National High Blood Pressure Education Programm (USA); Norwegian Society of Hypertension; Polish Society of Hypertension; Portuguese League against Hyper- tension; Senegal Foundation for Hypertension and Heart Disease Control; Slovene Hypertension Society; Southern African Hypertension Society; Spanish League against Hypertension; Swedish Hypertension Society; Swiss Association against High Blood Pressure; Uruguayan Hypertension League; Venezuelan Foundation against Hypertension and Ischaemic Heart Disease; Zaire League against Arterial Hypertension. Reprint No. 5331 Introduction The merits of non-pharmacological approaches in the management of hypertension have been considered in numerous reviews, meta-analyses, and reports (1, 2). The World Hypertension League also recently published consensus statements on three aspects of this subject - weight control (3), physical activity (4), and moderation of alcohol consumption (5). Although consensus on all aspects of non-pharmaco- logical interventions has not yet been attained, the evidence in support of such an approach has been considered convincing (6). This article describes the impact of non-pharmacological interventions which may reinforce or augment the effects of drug treat- ment for hypertension, thereby potentially reducing the patients' drug requirements. Manoeuvres like potassium or calcium loading, which have not been assessed in relation to interactions with antihyperten- sive drugs, are not discussed. Salt restriction Salt restriction and the response to antihypertensive drugs has been the topic of a recent review (7). Salt restriction generally results in an additional decrease in blood pressure in the presence of most (but not all) antihypertensive agents. Thus, it augmented the hypotensive effect of chlorthalidone in two investi- gations, but not that of hydrochlorothiazide in another (8-11), and increased the effect of beta- blockers in three studies (7, 10, 12). Angiotensin converting enzyme (ACE) inhibitors lowered blood pressure even more effectively when dietary salt intake was reduced (13). Bulletin of the World Health Organization, 70 (6): 685-690 (1992) 685 World Hypertension League The joint effect of ACE inhibitors and salt res- triction was recently investigated in detail by Singer et al. (14). They performed a double-blind crossover study of the effects of moderate salt reduction in 21 patients with essential hypertension who were al- ready being treated with a combination of a convert- ing enzyme inhibitor and a diuretic. After one month of captopril (50 mg twice daily) and hydrochloro- thiazide (25 mg once daily) therapy at their usual level of salt intake, the subjects were instructed in a reduced salt intake (80-100 mmol/day sodium). Their supine blood pressure was 147/96 ± 5/2 mmHg prior to salt reduction. After achieving a reduced salt intake, the subjects were randomized to "slow sodium" capsules or placebo. The study showed an added effect of salt restriction on the mean arterial blood pressure, which was reduced by 5-7%. The reduction was correlated with the reduced salt in- take. The authors suggested that the blockade of the renin-angiotensin system by captopril may augment the effects of salt restriction, since presumably the increases in angiotensin II due to salt restriction are thereby avoided. In his commentary on this paper, Morgan (15) pointed out that salt restriction is a desirable alternative to increasing the doses of diuretics. Increased diuretics could lead to further metabolic side-effects such as loss of potassium, a decrease in magnesium levels, higher uric acid values, and additional cholesterol elevations. He sug- gested that a combination of low-dose thiazide ther- apy and moderate salt restriction should be ideal in achieving the maximum blood pressure reduction with the minimum of side-effects. Morgan suggest- ed that moderate salt restriction (50-100 mmol/d) is readily attained by not adding salt at the table or in cooking, using low-salt bread and cereals, and avoid- ing high-salt foods (fast foods, processed meats). Weinberger et al. (16) reduced the dietary salt intake of 100 patients undergoing pharmacological treatment for hypertension. Their blood pressures were well controlled (140/90 mmHg) with medica- tions, which included a thiazide diuretic in all patients. Beta-blockers, calcium antagonists, conver- ting enzyme inhibitors and sympatholytics were also employed. This open study was designed to explore the feasibility of long-term salt intake reduction in an American clinic population. The subjects were given chloride titrator strips so that they could assess their level of adherence to the diet in first morning-voided urine samples (17). The educational methodology of this study has been published separately (18). Half of the subjects attaining sodium excretions of less than 80 mmol/d sufficiently reduced their blood pressures to permit a reduction of their medi- cations. Since only half of these motivated subjects were able to achieve adherence despite the presence of a dedicated team to help them to do so, it appears that a reduced salt intake may not be as easily at- tained as Morgan (15) and others (14) have suggest- ed. A few rules with respect to food preparation are generally not enough; professional help is needed. Finally, there is evidence that a reduced salt intake has no additive effect with calcium antagonist drug therapy (19, 20). Weight control This subject was discussed in detail in an earlier consensus statement (3). Briefly, obesity may account for as much as one-third of all hypertension (21). That weight loss results in decrease of blood pres- sure in overweight hypertensive persons, indepen- dently of other variables such as salt intake, has been clearly demonstrated (22). Further weight loss causes salt-sensitive individuals to become less salt sensitive (23). McMahon et al. (24) have also shown that weight loss lowers blood pressure in over- weight hypertensive patients compared to controls, while at the same time decreasing their cardiovascular risk. In a 21-week study, weight loss reduced blood pressure as effectively as metoprolol; at the same time HDL-cholesterol increased and total cholesterol decreased. Metoprolol also decreased blood pres- sure, but increased total cholesterol and decreased HDL-cholesterol. As with salt reduction, adherence is a major problem. Eliahou et al. reported on weight loss as a management strategy in over 200 drug-treated subjects visiting a community clinic (25). They reported that 20% of their patients did not complete the treatment programme and that 40% did not adhere to the dietary regimen. However, in the 60% of patients remaining, more than two-thirds achieved normal blood pressure with a loss of half their excess weight, even if they remained above ideal body weight. Oberman et al. (26) conducted a multicentre trial of pharmacological and non-pharmacological therapies in 692 hypertensive patients, who were 110-160% above ideal body weight. The diet inter- ventions were: usual, low salt plus high potassium, and weight loss. The drug interventions were: placebo, chlorthalidone, and atenolol. Nine diet/drug combinations were studied. After six months, the chlorthalidone plus usual diet group had a poorer car- diovascular risk profile, despite decreased blood pres- sure, because of undesirable lipid changes. All other groups showed favourable cardiovascular risk profiles; the best ones were identified in groups combining drug therapy plus weight loss. The authors concluded that weight loss was the most important adjunctive treatment in reducing overall cardiovascular risk. WHO Bulletin OMS. Vol 70 1992686 Non-pharmacological interventions and drug treatment of hypertension Alcohol consumption This topic was also recently reviewed in detail (5). Numerous epidemiological studies suggest that excessive alcohol intake increases blood pressure in a dose-related fashion in both men and women. A direct pressor effect of regular alcohol consumption was demonstrated even in normotensive subjects by Puddey et al. (28). This group also (29) examined the interaction between regular alcohol intake and pharmacological therapy in hypertensive subjects. Forty-four men with treated essential hypertension who were moderate to heavy drinkers took part in a randomized, controlled, crossover trial of the effects of alcohol intake on blood pressure. Usual antihyper- tensive treatment was maintained throughout six weeks of normal drinking and six weeks of drinking only a low-alcohol beer. Estimated alcohol consump- tion decreased from 452 to 64 ml per week in these subjects. Systolic and diastolic blood pressures were 5 and 3 mmHg lower, respectively, at the end of the low-alcohol, compared to the high-alcohol period. Regression analysis suggested that reduction in alco- hol intake contributed to the fall in both systolic and diastolic blood pressures independently of changes in body weight. The authors concluded that curtailing alcohol intake may reduce the need for antihyperten- sive drugs (29). Reduction in heavy drinking should be an integral part of any comprehensive manage- ment programme for hypertensive patients. Exercise The role of physical exercise in the management of hypertension was the subject of a World Hyper- tension League statement in 1991 (4). Regular vigor- ous exercise is associated with a reduced risk for hypertension (30). The effect of exercise is indepen- dent of body weight, alcohol ingestion, or electrolyte intake (4). Recently an epidemiological investigation of physical activity of male college alumni from the University of Pennsylvania (32) has shown that physical activity is associated with reduced occur- rence of non-insulin-dependent diabetes mellitus (NIDDM). NIDDM is closely associated with hyper- tension, and the occurrence of both greatly adds to the risk of cardiovascular events. Interestingly, the protective effects of exercise were greatest for those at the highest risk for developing the disease. Since exercise improves glucose tolerence, lowers insulin levels, increases peripheral sensitivity to insulin, and decreases sympathetic tone, mechanisms are at hand which may explain these findings. These mecha- nisms are of interest in terms of the association recently identified between disturbed carbohydrate metabolism and hypertension, even in hypertensive patients who have normal glucose tolerance (33). This association is receiving considerable attention (34). A number of well-performed prospective, ran- domized, clinical trials indicate that exercise de- creases blood pressure in patients with hypertension to a similar degree as do salt restriction or weight loss (35-37). The effect of exercise is independent of weight loss. However, both are desirable. Exercise results in weight loss accompanied by preserved lean body mass, while diet alone decreases both body weight and lean body mass. Numerous studies on the interaction between aerobic exercise and antihypertensive agents have been performed; however, these have generally examined the effects of antihypertensive agents on exercise performance (31). An exception is the study by Ades and associates (38). These investigators ran- domized hypertensive patients into three groups. A non-drug-treated, but exercised control group was compared to a propranolol-treated, exercised group and a metoprolol-treated, exercised group. Exercise decreased mean blood pressure by about 8-10 mmHg in the non-drug-treated group and in the metoprolol-treated group, but not in the propranolol- treated group. It is possible that the antagonism of B-2 receptors may interfere with blood pressure- lowering effects of exercise. However, the study also indicates that exercise may lower blood pressure in patients treated for hypertension. In conclusion, exercise both lowers blood pres- sure and has independent, desirable effects on body weight, plasma lipid levels, and carbohydrate toler- ance. Exercise should be prescribed as part of every non-pharmacological regimen. A tailored programme of walking, avoidance of elevators, and leisure-time activity should be encouraged. If ischaemic heart disease or left ventricular hypertrophy is present, it is advisable to begin exercise with a cautious, super- vised programme. Potassium A large body of evidence suggests that potassium intake is inversely associated with blood pressure and that an increase in potassium may lower the pressure in patients with hypertension. Numerous trials have been conducted to address the latter issue. These trials were recently subjected to a meta- analysis by Cappuccio & McGregor (39). A total of 19 clinical trials were examined. According to their calculations, systolic blood pressure was lowered by about 6 mmHg and diastolic blood pressure by about 4 mmHg with potassium supplements. The magni- WHO Bulletin OMS. Vol 70 1992 687 World Hypertension League tude of the blood-pressure-lowering effect was great- er in patients with higher blood pressures and appear- ed to be more pronounced the longer the duration of the supplementation. The authors conclude that potassium supplementation could allow a reduction in the need for antihypertensive medication in many patients. Grimm et al. (40) conducted a randomized, placebo-controlled, double-blind clinical trial of potassium chloride supplementation in 287 hyperten- sive men treated with drugs, to determine whether or not potassium chloride could reduce the drug requirements. A total of 142 men were given potas- sium chloride and 145 were given placebo in addi- tion to a low sodium diet. Medication was then withdrawn and they were followed for a period of 2.2 years. Seventy-nine participants in each group required restitution of antihypertensive medication according to strict indications defined by protocol. No significant differences in systolic or diastolic blood pressure were observed between the groups. The authors concluded that potassium chloride supplementation did not reduce the need for anti- hypertensive medication in hypertensive men on a salt-restricted diet. In their commentary on this paper, Kaplan & Ram (41) pointed out that the study by Grimm et al. (40) involved only white men, and that benefits of potassium supplementation may be greater in black as compared to white subjects. Further, they argued that preventing an increase in blood pressure may differ from initiating a decrease. In addition, a low- salt diet would be expected to minimize any effects of potassium, since potassium salts seem to exert their effects by initiating natriuresis. Finally, they pointed out that evidence exists showing desirable vascular protective effects of a high potassium intake that are independent of effects on blood pressure. All of these points may be important. However, supplements are expensive and may occasionally be dangerous. Whether or not non-chloride-containing potassium salts exert the same effect as potassium chloride, particularly with respect to natriuresis, is not clear. Overlack (42) was not encouraged by the blood-pressure-lowering effects of potassium com- bined with non-chloride anions. On the other hand, Siani et al. (43) reported highly impressive effects on blood pressure resulting from a high-potassium diet consisting largely of steamed, rather than boiled vegetables. This regimen allowed a reduction of the patients' medication. The putative mechanisms of such a manoeuvre have recently been addressed by Krishna & Kapoor (44). These investigators perform- ed a metabolic study on 12 patients with hyperten- sion who received a fixed diet except for potassium (in the food), which was offered at either 16 or 96 mmol/d. Low potassium intake was associated with a 6 mmHg increase in mean blood pressure, a decrease in sodium excretion, and a decrease in renin and aldosterone, while the arginine, vasopressin, and atrial natriuretic peptide values were not affected. Thus, potassium and sodium intakes and excretion appear irrevocably connected. While potassium supplementation is not a non- pharmacological intervention, it provides cues regarding the importance of a natural diet high in potassium. A high-potassium diet is likely to contain more fresh fruits and vegetables, thereby favour- ably influencing salt, fat, and probably also calorie intake. Such a diet is as a matter of principle prefer- able to one low in potassium, consisting largely of processed foods from which the potassium has been removed. Replacing the potassium afterwards with potassium chloride is like placing the cart before the horse. Conclusion In summary, any antihypertensive drug therapy should be accompanied by intensive non-pharmaco- logical treatment. Weight control, alcohol modera- tion, and regular exercise should be immediate goals. Ample fruits and vegetables, fresh rather than pro- cessed foods in moderate amounts, and attention to reduced fat consumption and avoidance of added salt intake will result in a diet that would satisfy most if not all advocates of dietary management. Concentra- tion on any single electrolyte appears to be a narrow approach; instead a holistic one is appropriate (45). Non-pharmacological approaches lower the drug requirements in patients with hypertension. Such approaches may enable some patients to discontinue drug treatment indefinitely (46). A decreased salt intake to a sodium value of about 80 mmol/d lowers blood pressure in patients treated with diuretics, beta-blockers, converting enzyme antagonists and sympatholytics, but apparently not in those patients treated solely with calcium antagonists. A reduced medication requirement is seen in about half of patients, suggesting that salt-sensitivity plays a role as well. Weight loss is an effective adjunct therapy and simultaneously improves other cardiovascular risk factors. In a recent multicentre comparison of various diets and drug regimens, weight loss increased the effects of chlorthalidone, while a low- sodium plus high-potassium diet did not (47). Reduc- ing the alcohol intake of hypertensive heavy drinkers significantly lowers their blood pressures and drug requirements. Regular exercise is an effective non- pharmacological treatment, which simultaneously fosters weight loss and improves risk factors. Its WHO Bulletin OMS. Vol 70 1992688 Non-pharmacological interventions and drug treatment of hypertension effects may be blunted by some pharmacological treat- ments. Potassium chloride supplements do not decrease medication requirements in drug-treated patients who are ingesting a low-salt diet. Arguments in favour of a natural high-potassium diet are never- theless compelling. However, the first step in non-pharmacological treatment should concentrate on whether or not the patient smokes cigarettes. This is because stopping smoking is more important for the patient's welfare than either pharmacological or non-pharmacological treatment. Although smoking on its own does not contribute to the elevation of blood pressure, the cardiovascular risk of the smoking hypertensive patient is so high that the importance of stopping smoking overshadows all other non-pharmacological interventions. This, however, should not detract attention from the drug-sparing potential of dietary measures and exercise. References 1. The 1984 Report of the Joint National Committee on Detection, Evaluation, and Treatment of High Blood Pressure. Arch. intern, med., 144: 1045-1057 (1984). 2. Chockalingam, A. et al. Recommendations of the Canadian Consensus Conference on Non-Pharma- cological Approaches to the Management of High Blood Pressure, 21-23 March 1989, Halifax, Nova Scotia. Can. Med. Assoc. j., 142: 1397-1409 (1 990). 3. World Hypertension League. Weight control in the management of hypertension. Bull. Wld Hlth Org., 67: 245-252 (1989). 4. World Hypertension League. Physical exercise in the management of hypertension. J. hypertens., 9: 283-287 (1991) and Bull. Wld Hlth Org., 69: 149-153 (1991). 5. World Hypertension League. Alcohol and hyper- tension - implications for management. J. human hypertens., 5: 227-232 (1991) and Bull. Wld Hlth Org., 69: 377-382 (1991). 6. McCarron, D.A. et al. Education program on non- pharmacologic management of hypertension. New York, National Kidney Foundation Inc., 1990. 7. Luft, F.C. & Weinberger, M.H. Review of salt res- triction and the response to antihypertensive drugs. Hypertension, 11(suppl. 1): 1-229-1-232 (1988). 8. Erwteman, T.M. et al. Betablockade, diuretics, and salt restriction for the management of mild hyperten- sion: a randomized, double-blind trial. Br. med. j., 289: 406-409 (1984). 9. Ram, C.V. & Kaplan, N.M. Moderate sodium res- triction and various diuretics in the treatment of hypertension. Arch. intern. med., 141: 1015-1019 (1981). 10. Carney, S.L. et al. Effect of dietary sodium restric- tion on patients receiving antihypertensive medica- tion. Clin. exp. hypertens., A6: 1095-1105 (1984). 11. Owens, C.J. & Brackett, N.C. Role of sodium in- take in the antihypertensive effect of propranolol. South med. j., 71: 43-46 (1978). 12. Pollavini, G. et al. Effects of moderate salt restric- tion in hypertensive patients treated with oxprenolol or chlorthalidone. Inter. j. clin. pharmacol. therap. toxicol., 22: 451-455 (1984). 13. Hollenberg, N.K. et al. Sodium intake and renal responses to captopril in normal man and in essen- tial hypertension. Kidney international, 20: 240-245 (1981). 14. Singer, D.R.J. et al. Sodium restriction in hyperten- sive patients treated with a converting enzyme inhi- bitor and a thiazide. Hypertension, 17: 798-803 (1991). 15. Morgan, T. Interaction of pharmacological and non- pharmacological therapy. Hypertension, 17: 804-805 (1991). 16. Weinberger, M.H. et al. Dietary sodium restriction as adjunctive treatment of hypertension. J. Am. Med. Assoc., 259: 2561-2565 (1988). 17. Luft, F.C. et al. The utility of overnight urine collec- tions in assessing compliance with a low sodium intake diet. J. Am. Med. Assoc., 249: 1764-1768 (1983). 18. Cohen, S.J. et al. The effect of a household partner and home urine monitoring on adherence to a sodium restricted diet. Soc. sci. med., 32: 1057-1061 (1991). 19. Morgan, T.O. & Anderson, A. Interaction of slow- channel calcium-blocking drugs with sodium restric- tion, diuretics and converting enzyme inhibitors. J. hypertens., 6(suppl. 4): S-652-S-654 (1988). 20. Morgan, T.O. et al. Paradoxical effect of sodium restriction on blood pressure in people on slow- channel calcium-blocking drugs. Lancet, 1: 793 (1986). 21. MacMahon, S.W. et al. Obesity, alcohol consump- tion and blood pressure in Australian men and women. The National Heart Foundation of Australia Risk Factor Prevalence Study. J. hypertens., 2: 85-91 (1984). 22. Reisin, E. et al. Effect of weight loss without salt restriction in the reduction of blood pressure in over- weight hypertensive subjects. New Engl. j. med., 298: 1-6 (1978). 23. Rocchini, A.P. et al. The effect on weight loss on the sensitivity of blood pressure to sodium in obese adolescents. New Engl. j. med., 321: 580-585 (1989). 24. MacMahon, S.W. et al. A randomized controlled trial of weight reduction and metoprolol in the treat- ment of hypertension in young overweight patients. Cin. exp. pharm. physiol., 12: 267-271 (1985). 25. Eliahou, H.E. et al. Body weight reduction neces- sary to attain normotension in the overweight hyper- tensive patient. Inter. j. obes., 5(suppl. 1): 157-163 (1981). 26. Oberman, A. et al. Pharmacologic and nutritional treatment of mild hypertension: changes in cardio- vascular risk status. Ann. intern. med., 112: 89-95 (1 990). 27. MacMahon, S. Alcohol consumption and hyper- tension. Hypertension, 9: 111-121 (1987). WHO Bulletin OMS. Vol 70 1992 689 World Hypertension League 28. Puddey, I.B. et al. Evidence for a direct effect of alcohol consumption on blood pressure in normoten- sive men: a randomized controlled trial. Hyper- tension, 7: 707-713 (1985). 29. Puddey, I.B. et al. Regular alcohol use raises blood pressure in treated hypertensive subjects. Lancet, 1: 647-651 (1987). 30. Paffenbarger, R.S. et al. Physical activity and inci- dence of hypertension in college alumni. Am. j. epi- demiol., 117: 245-257 (1983). 31. Luft, F.C. Hypertension and exercise. In: Grana, W.A. et al., ed. Advances in sports medicine and fit- ness, vol. 2. Chicago, Year Book Med. PubI., 1989, pp. 119-135. 32. Helmrich, S.P. et al. Physical activity and reduced occurrence of non-insulin-dependent diabetes melli- tus. New Engl. j. med., 325: 147-152 (1991). 33. Ferrannini, E. et al. Insulin resistance in essential hypertension. New Engl. j. med., 317: 350-357 (1987). 34. Donahue, R.P. et al. Hyperinsulinemia and elevated blood pressure: cause, confounder, or coincidence? Am. j. epidemiol., 132: 827-836 (1990). 35. Nelson, L. et al. Effect on changing levels of physi- cal activity on blood pressure and hemodynamics in essential hypertension. Lancet, 2: 473-476 (1986). 36. Pagani, M. et al. Changes in autonomic regulation induced by physical training in mild hypertension. Hypertension, 12: 600-610 (1988). 37. Hagberg, J.M. et al. Effect of exercise training on the blood pressure and hemodynamics of adoles- cent hypertensives. Am. j. cardiol., 52: 763-768 (1983). 38. Ades, P.A. et al. Hypertension, exercise, and beta- adrenergic blockade. Ann. intern. med., 109: 629-634 (1988). 39. Cappuccio, F.P. & MacGregor, G.A. Does potas- sium supplementation lower blood pressure? A meta-analysis of published trials. J. hypertens., 9: 465-473 (1991). 40. Grimm, R.H. et al. The influence of oral potassium chloride on blood pressure in hypertensive men on a low-sodium diet. New Engl. j. med., 322: 569-574 (1 990). 41. Kaplan, N.M. & Ram, C.V.S. Potassium supple- ments for hypertension. New Engl. j. med., 322: 623-624 (1990). 42. Overlack, A. et al. The influence of oral potassium citrate/bicarbonate on blood pressure in essential hypertension during unrestricted salt intake. Klin, Wochenschr., 69(suppl. XXV): 79-83 (1991). 43. Siani, A. et al. Increasing dietary potassium intake reduces the need for antihypertensive medication [abstract]. Am. j. hypertens., 3: 11 OA (1990). 44. Krishna, G.G. & Kapoor, S.C. Potassium depletion exacerbates essential hypertension. Ann. intern. med., 115: 77-83 (1991). 45. Luft, F.C. & McCarron, D.A. Heterogeneity of hypertension: the diverse role of electrolyte intake. Annual rev. med., 42: 347-355 (1991). 46. Langford, H.G. et al. Dietary therapy slows the return of hypertension after stopping prolonged medication. J. Am. Med. Assoc., 253: 657-664 (1985). 47. Langford, H.G. et al. for the TAIM Research Group. Effect of drug and diet treatment of mild hypertension on diastolic blood pressure. Hyper- tension, 17: 210-217 (1991). 690 WHO Bulletin OMS. Vol 70 1992