053101 correspondence

The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne Editor’s note: Dr. Alderman has attended consulting 1. Sacks FM, Svetkey LP, Vollmer WM, et al. Effects on blood pressure of
reduced dietary sodium and the Dietary Approaches to Stop Hypertension
(DASH) diet. N Engl J Med 2001;344:3-10.
2. Graudal NA, Galloe AM, Garred P. Effects of sodium restriction on
blood pressure, renin, aldosterone, catecholamines, cholesterols, and tri-
glyceride: a meta-analysis. JAMA 1998;279:1388-91.
Dietary Sodium and Blood Pressure
3. He J, Ogden LG, Vupputuri S, Bazzano LA, Loria C, Whelton PK. Di-
etary sodium intake and subsequent risk of cardiovascular disease in over-
weight adults. JAMA 1999;282:2027-34.
To the Editor: The DASH–Sodium Trial (Jan. 4 issue)1 4. Alderman MH, Cohen H, Madhavan S. Dietary sodium intake and
has demonstrated that a reduction in sodium intake over mortality: the National Health and Nutrition Examination Survey a period of 30 days lowers blood pressure. This finding is consistent with what more than 100 previous short-term,randomized trials have convincingly shown. What is remark-able — and disappointing, given this demanding and cost- To the Editor: Sacks et al. concluded that their results in ly undertaking — is that the report was limited to the sin- the DASH–Sodium Trial “provide support for a more ag- gle selected favorable effect of salt on blood pressure, to gressive target for reduced sodium intake, in combination the exclusion of other important physiological effects. For with the use of the DASH (Dietary Approaches to Stop Hy- example, reducing the consumption of sodium also increas- pertension) diet, for the prevention and treatment of ele- es the plasma renin level by a factor of 3.6 and the aldos- vated blood-pressure levels.” That conclusion is misleading terone level by a factor of 3.2 — effects proportional to Given the characteristics of the study cohort, which are The relation between sodium intake and all physiological greatly skewed toward the expression of salt sensitivity, the phenomena is of great scientific interest. By contrast, pa- failure to provide the Journal’s readers with all the avail- tients, physicians, and public health advocates need to know able data must be addressed. Even the very limited data pro- how salt intake affects the quality and duration of life. The vided in Figures 1 and 2 of the article indicate that from scanty published data available are not encouraging to those the standpoint of clinical application, any effect of salt re- who preach salt restraint. An exception is the finding in a striction on blood pressure was limited to black women with subgroup of obese subjects (28 percent) in the National hypertension. Thus, the authors’ statement that these find- Health and Nutrition Examination Survey Epidemiologic ings are broadly applicable to the entire population is not Follow-up Study of a direct relation between salt intake true. To prove otherwise, the authors are obligated to report, and cardiovascular outcome. No such relation appeared in in a format similar to that of Figure 1, the analyses of sub- the nonobese majority (72 percent).3 Moreover, in the en- groups defined according to race, presence or absence of tire survey population, there was an inverse relation between hypertension, sex, age, and body-mass index.
sodium intake and cardiovascular morbidity.4 Those data are necessary to document the principal find- Perhaps in some people the hemodynamic benefits of ing of the study: for the vast majority of people, overall salt restriction will outweigh its hormonal and metabolic dietary improvements eliminate the effects of salt on blood hazards. However, at this point, in the absence of strong pressure. Earlier reports that adequate mineral intake from evidence of a consistent health effect, a single universal rec- dairy products, fruits, and vegetables is far more important ommendation for a reduction in dietary sodium reflects than salt in determining blood pressure have been con- firmed by this study. If the authors had provided graphsthat told the whole story, that conclusion would have been Unfortunately, as currently presented, these data only provide further justification to the 70 percent of recently Letters to the Editor are considered for publication (subject to editing and abridgment) provided they do not contain material that has been submitted or published elsewhere. Please note the following: •Your letter must be typewritten and triple-spaced. •Its text, not including references, must not exceed 250 words if it is in reference to a recent Journal article, or 400 words in all other cases (please provide a word count). •It must have no more than five references and one figure or table. •It must not be signed by any more than three authors. •Letters referring to a recent Journal article must be received within four weeks of its publication. •Please include your full address, telephone number, fax number, and e-mail address. •You may send us your letter by standard mail, fax, or e-mail. Our address: Letters to the Editor • New England Journal of Medicine • 10 Shattuck St. • Boston, MA 02115
Our fax numbers: 617-739-9864 and 617-734-4457
Our e-mail address: letters@nejm.org
We cannot acknowledge receipt of your letter, but we will notify you when we have made a decision about publication. We are unable to provide prepublication proofs. Financial associations or other possible conflicts of interest must be disclosed. Submission of a letter constitutes permission for the Massachusetts Medical Society, its licensees, and its assignees to use it in the Journal’s various print and electronic publications and in collections, revisions, and any other form or medium.
1716 · N Engl J Med, Vol. 344, No. 22 · May 31, 2001 · www.nejm.org
Downloaded from www.nejm.org at CILEA BIBLIOSAN on November 17, 2009 . Copyright 2001 Massachusetts Medical Society. All rights reserved. C O R R E S P O N D E N C E
surveyed Americans who said they do not trust the gov- apy shows a reduction in the risks of stroke and coronary ernment’s dietary recommendations.3 As others have put it, heart disease in relation to reductions in diastolic but not the salt-restriction advocates want to protect the public not systolic blood pressure.3 To arrive at the best estimates of the expected reductions in disease events that would resultfrom changes in the diets of various subgroups — especially persons without hypertension — information is required about the effect on diastolic blood pressure of lowering di- etary sodium in the DASH and control diets for subgroupsdefined by ethnic background, presence or absence of hy- Editor’s note: Dr. McCarron is a member of the Medical Advisory Board of the Salt Institute.
The most recent meta-analysis of clinical trials of sodium reduction reported no significant effect of sodium reduc- 1. McCarron DA, Morris CD, Cole C. Dietary calcium in human hyper-
tion on diastolic blood pressure in persons without hyper- 2. McCarron DA, Morris CD, Henry HJ, Stanton JL. Blood pressure and
nutrient intake in the United States. Science 1984;224:1392-8.
3. Patterson RE, Satia JA, Kristal AR , Neuhouser ML, Drewnowski A. Is
there a consumer backlash against the diet and health message? J Am Diet Assoc 2001;101:37-41.
4. Freedman DA, Petitti DB. Salt and blood pressure: conventional wis-
dom reconsidered. Evaluation Review (in press).
To the Editor: The continuing controversy about general advice to reduce dietary sodium has arisen in part because the effect of sodium may be limited to subgroups of thepopulation who are salt-sensitive. In their discussion of the 1. Stamler J, Stamler R , Neaton JD. Blood pressure, systolic and diastolic,
effect of sodium reduction in persons without hyperten- and cardiovascular risks. Arch Intern Med 1993;153:598-615.
2. MacMahon S, Peto R , Cutler J, et al. Blood pressure, stroke, and cor-
sion, Sacks et al. highlight the reductions in systolic blood onary heart disease. I. Prolonged differences in blood pressure: prospective pressure that occurred in subjects on the control diet. Figure observational studies corrected for the regression dilution bias. Lancet 2 of the article shows that the reduction in sodium intake in subjects on the DASH diet had a significant effect on 3. Collins R , Peto R. Antihypertensive drug therapy: effects on stroke and
coronary heart disease. In: Swales JE, ed. Textbook of hypertension. Ox-
systolic blood pressure in persons without hypertension only ford, England: Blackwell Scientific, 1994:1156-64.
if they were black and only when they switched from the 4. Graudal NA, Galloe AM, Garred P. Effects of sodium restriction on
blood pressure, renin, aldosterone, catecholamines, cholesterols, and tri- The risk of coronary disease increases as the diastolic glyceride: a meta-analysis. JAMA 1998;279:1388-91.
blood pressure increases within quintiles of systolic bloodpressure.1 A meta-analysis of observational studies of blood To the Editor: In their report of the 90-day DASH– pressure, stroke, and coronary disease2 presents data accord- Sodium Trial, Sacks et al. do not set their findings in the ing to diastolic but not systolic blood pressure. A meta- context of longer-term trials in which sodium intake has analysis of randomized trials of antihypertensive-drug ther- been manipulated to reduce blood pressure. We have iden- TABLE 1. RESULTS OF LARGE RANDOMIZED, CONTROLLED TRIALS IN WHICH
SODIUM INTAKE WAS LOWERED TO REDUCE BLOOD PRESSURE.* TRIAL AND YEAR
SUBJECTS
SYSTOLIC BLOOD PRESSURE
*For the first three trials, the net mean change in systolic blood pressure was calculated by sub- tracting the mean change in the control group from the mean change in the experimental group. Forthe DASH–Sodium Trial, the data presented represent the mean change for the 204 participants inthe control group from the high-sodium phase to the low-sodium phase of the control diet. CI de-notes confidence interval, HPT Hypertension Prevention Trial, TOHP Trial of Hypertension Preven-tion, and DASH Dietary Approaches to Stop Hypertension. In the 1998 Trial of NonpharmacologicInterventions in the Elderly, involving 681 subjects, the hazard ratios for the composite end point ofhypertension, restarting medication, or a cardiovascular event by 30 months (in the sodium-reductiongroup vs. the control group) were 0.60 (95 percent confidence interval, 0.45 to 0.80) in obese sub-jects and 0.75 (95 percent confidence interval, 0.59 to 0.95) in nonobese subjects.
N Engl J Med, Vol. 344, No. 22 · May 31, 2001 · www.nejm.org · 1717
Downloaded from www.nejm.org at CILEA BIBLIOSAN on November 17, 2009 . Copyright 2001 Massachusetts Medical Society. All rights reserved. The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne tified 10 randomized, controlled trials that lasted six months cause people have serious difficulty finding enough low- or longer. Four trials, all conducted in the United States, sodium foods to enable them to adhere fully to reduced- were similar in size to the DASH–Sodium Trial, having at sodium diets. We hope that the food industry will respond least 200 participants (Table 1, previous page).
to the DASH–Sodium findings as a business opportunity.
In three of these trials, involving adults not receiving med- We are surprised by Alderman’s assertion that reduced ication who had high-normal blood pressure,1-3 the differ- sodium intake, especially to 65 mmol per day, is harmful.
ence between the mean reduction in systolic blood pressure Reduced sodium does raise the plasma renin level, but so in the experimental group and that in the control group was does diuretic therapy, which prevents cardiovascular disease.
modest after 6 months of intervention and became mini- Moreover, in a large sample of the general population, the mal after 12 months. In the fourth trial, the Trial of Non- plasma renin level was not associated with cardiovascular dis- pharmacologic Interventions in the Elderly,4 however, there ease.1 Experience with antihypertensive drugs shows that was observable but limited success in weaning seniors off any reduction in blood pressure reduces cardiovascular dis- antihypertensive medication by means of a low-sodium diet.
ease. Mortality trials are not required for drug approval.
In all four trials, there was difficulty in maintaining so- Should sodium reduction be different? Finally, the epide- dium reduction at the targeted level over the long term, miologic study by Alderman et al.2 shows an increase in the even though the interventions were intensive. This may ex- rate of death from cardiovascular disease with higher salt plain the minimal efficacy, although an alternative hypothesis intake when the analysis is appropriately adjusted for the is that baroreceptor and renal homeostasis through renin– total caloric intake of the subjects.
angiotensin mechanisms restored blood pressure to its pre- We assure McCarron and readers that the DASH–Sodium intervention level, despite the reduction in sodium.5 Thus, findings are qualitatively the same among all subgroups but it is improbable that the reductions in blood pressure as- differ quantitatively, as we showed in representative exam- cribed by Sacks and colleagues to the manipulation of so- ples. We plan to publish the full subgroup results soon.
dium intake (as distinct from the reductions that were due The substantial reductions in blood pressure give additional to the DASH diet) could be sustained over the long term in strength to the long-standing dietary guidelines of public adults not receiving medication, either in a clinical situation or in the general population. Nevertheless, the issue of Petitti and Freedman focus on a subgroup effect that did whether low sodium intake from childhood promotes life- not attain statistical significance, while ignoring the consis- long normotension remains to be resolved.
tency of the direction of the results for all subgroups, whichvary only in magnitude. Moreover, a substantial proportion of white and normotensive persons are indeed sensitive to salt. Systolic blood pressure was prespecified as the trial’s primary outcome because it is more closely linked than di-astolic blood pressure to cardiovascular disease.4 We agree with the view of Bartlett et al. that difficulty in achieving long-term adherence to low-sodium diets limits the long-term reduction in blood pressure. In the outstand-ing trial of MacGregor et al.,5 the reduction of sodium in- take from 200 to 50 mmol per day in a highly motivatedgroup of subjects with hypertension reduced systolic blood pressure by 18 mm Hg after one year — the same as thereduction after one month. There is no indication in the 1. The Hypertension Prevention Trial Research Group. The Hypertension
Prevention Trial: three-year effects of dietary changes on blood pressure. Arch Intern Med 1990;150:153-62.
2. The effects of nonpharmacologic interventions on blood pressure of
persons with high normal levels: results of the Trials of Hypertension Pre- vention, Phase I. JAMA 1992;267:1213-20. [Erratum, JAMA 1992;267:
2330.]
3. The Trials of Hypertension Prevention Collaborative Research Group.
Effects of weight loss and sodium reduction intervention on blood pressure National Heart, Lung, and Blood Institute and hypertension incidence in overweight people with high-normal blood pressure: the Trials of Hypertension Prevention, Phase II. Arch Intern Med
1997;157:657-67.
4. Whelton PK, Appel LJ, Espeland MA, et al. Sodium reduction and
weight loss in the treatment of hypertension in older persons: a random- ized controlled trial of nonpharmacologic interventions in the elderly (TONE). JAMA 1998;279:839-46.
5. Navar LG. The kidney in blood pressure regulation and development
of hypertension. Med Clin North Am 1997;81:1165-98.
1. Meade TW, Cooper JA, Peart WS. Plasma renin activity and ischemic
heart disease. N Engl J Med 1993;329:616-9.
To the Editor: The contribution of the DASH–Sodium 2. Alderman MH, Cohen H, Madhavan S. Dietary sodium intake and
Trial to the salt issue is data on the magnitude of the re- mortality: the National Health and Nutrition Examination Survey duction in blood pressure related to a reduction in sodium.
(NHANES I). Lancet 1998;351:781-5.
3. Stamler J, Neaton JD, Wentworth DN. Blood pressure (systolic and di-
In our view, the magnitude is greater in the DASH–Sodium astolic) and risk of fatal coronary heart disease. Hypertension 1989;13: feeding study than has been reported in meta-analyses be- 1718 · N Engl J Med, Vol. 344, No. 22 · May 31, 2001 · www.nejm.org
Downloaded from www.nejm.org at CILEA BIBLIOSAN on November 17, 2009 . Copyright 2001 Massachusetts Medical Society. All rights reserved. C O R R E S P O N D E N C E
4. Weinberger MH, Miller JZ, Luft FC, Grim CE, Fineberg NS. Defini-
1. Committee on Infectious Diseases. Recommended childhood immuni-
tions and characteristics of sodium sensitivity and blood pressure resist- zation schedule: United States, January-December 2001. Pediatrics 2001; ance. Hypertension 1986;8:Suppl II:II-127–II-134.
5. MacGregor GA, Markandu ND, Sagnella GA, Singer DR , Cappuccio
2. Black S, Shinefield H, Fireman B, et al. Efficacy, safety and immunoge-
FP. Double-blind study of three sodium intakes and long-term effects of nicity of heptavalent pneumococcal conjugate vaccine in children. Pediatr sodium restriction in essential hypertension. Lancet 1989;2:1244-7.
A Pneumococcal Conjugate Vaccine and Acute
To the Editor: Given the poor overall efficacy of the Otitis Media
pneumococcal conjugate vaccine in the study by Eskola etal., the authors’ conclusions are a bit too optimistic. Accord- To the Editor: In the report by Eskola et al. (Feb. 8 is- ing to the protocol, all infants received four vaccinations, sue)1 of the efficacy of a pneumococcal conjugate vaccine which were certainly not painless and which led to the pre- against otitis media, the data argue strongly against their vention of only 6 percent of all cases of acute otitis media.
conclusion that “the effect of the pneumococcal conjugate For the moment, more could be gained by changing vaccine can be substantial.” In the case of 94 percent of our attitude toward acute otitis media. In the Netherlands, those immunized with the pneumococcal conjugate vaccine, acute otitis media is seen as a self-limiting disease. Often, otitis media would not have been prevented. The study parents do not take their children to the doctor for this also substantiates one of the greatest feared limits of the common childhood disease. The efficacy of antibiotics for heptavalent pneumococcal conjugate vaccine — that the this disease is moderate,1 even in the youngest children.2 frequency of disease would remain the same as a result of Educating doctors and parents about acute otitis media increases in the rates of infection with all the pneumococ- will lead to a decrease in antibiotic prescriptions for this cal serotypes that were not included in the vaccine (a phe- illness, as has been shown in a recent trial.3 nomenon sometimes referred to as serotype replacement).
ROGER A.M.J. DAMOISEAUX, M.D., PH.D.
The data of Eskola et al. demonstrate a 33 percent increase in the incidence of otitis media from pneumococcal sero- types that were not included in the vaccine.
1. Del Mar C, Glasziou P, Hayem M. Are antibiotics indicated as initial
treatment for children with acute otitis media? A meta-analysis. BMJ 1997;314:1526-9.
2. Damoiseaux RAMJ, van Balen FAM, Hoes AW, Verheij TJ, de Melker
RA. Primary care based randomised, double blind trial of amoxicillin ver-sus placebo for acute otitis media in children aged under 2 years. BMJ 1. Eskola J, Kilpi T, Palmu A, et al. Efficacy of a pneumococcal conjugate
vaccine against acute otitis media. N Engl J Med 2001;344:403-9.
3. Little P, Gould C, Williamson I, Moore M, Warner G, Dunleavy J.
Pragmatic randomised controlled trial of two prescribing strategies for
childhood acute otitis media. BMJ 2001;322:336-42.
To the Editor: Eskola et al. state that “up to 1.2 million of the 20 million yearly episodes of acute otitis media in To the Editor: The vaccine manufacturer’s vice president the United States could theoretically be prevented if the and contractors conclude that the new pneumococcal vac- [pneumococcal conjugate] vaccine were widely used.” This cine is “efficacious in the prevention of acute otitis me- theoretical benefit must not be used as an official justifi- dia.” Given the 95 percent confidence interval of –4 to 16 percent, the data presented by Eskola et al. do not support The vaccine was officially recommended for use in all this conclusion. As the authors admit, “the negative number children on the basis of its efficacy against invasive diseas- indicates” that the treated group could have had more ep- es.1 The reported relative reduction in the risk of episodes of isodes than the controls. These same data — 1251 episodes 6 percent achieved by vaccination still leaves the total risk in the vaccine group as compared with 1345 episodes in high enough for many immunized children to have acute oti- the control group, a clinically and statistically insignificant tis media. This factor could lessen the credibility of the rec- difference in the rate of 3.6 percent (absolute benefit) — were presented to an advisory committee of the Food and A more powerful trial2 reported a reduction in acute Drug Administration (FDA) in November 1999. Therefore, otitis media of up to 22.8 percent in one subgroup of chil- the FDA rejected the use of this vaccine for otitis media.
dren with frequent episodes. Since 14 percent of children Moreover, the authors’ use of the number of episodes in- account for approximately 50 percent of the episodes of stead of the number of patients as the experimental unit is acute otitis media, the number of episodes could be signif- incorrect because it inflates the outcome.
icantly reduced by the selection of appropriate candidates The most interesting results concern the pneumococcal for immunization. In countries that will not adopt the U.S.
ecology. In a short time span, the predicted serotype replace- recommendation, selective vaccination of high-risk sub- ment1 (also observed with other bacterial vaccines2-4) was groups is a better way to keep the public trust.
realized. Eskola et al. report that the “number of episodesdue to all other [pneumococcal] serotypes increased by 33 percent.” With this clear warning sign, it is a perilous ex- periment in microbial ecology to push the use of the hep- University of Pittsburgh School of Medicine N Engl J Med, Vol. 344, No. 22 · May 31, 2001 · www.nejm.org · 1719
Downloaded from www.nejm.org at CILEA BIBLIOSAN on November 17, 2009 . Copyright 2001 Massachusetts Medical Society. All rights reserved. The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne 1. Spratt BG, Greenwood BM. Prevention of pneumococcal disease by
first ones (i.e., counting only one episode per child). This vaccination: does serotype replacement matter? Lancet 2000;356:1210-1.
approach did not inflate the outcome, as shown by the effi- 2. Perdue DG, Bulkow LR , Gellin BG, et al. Invasive Haemophilus influ-
enzae disease in Alaskan residents aged 10 years and older before and after
cacy rates against first as compared with all episodes of oti- infant vaccination programs. JAMA 2000;283:3089-94.
tis media (52 percent vs. 57 percent for infections caused by 3. Ramsay ME, Andrews N, Kaczmarski EB, Miller E. Efficacy of menin-
serotypes included in the vaccine and 12 percent vs. 6 per- gococcal serogroup C conjugate vaccine in teenagers and toddlers in Eng- cent for episodes from any cause). The reduction of 12 land. Lancet 2001;357:195-6.
4. Lipsitch M. Bacterial vaccines and serotype replacement: lessons from
percent is statistically significant (95 percent confidence Haemophilus influenzae and prospects for Streptococcus pneumoniae. We share the concern about the increase in otitis media caused by the serotypes that were not included in the vac-cine among those who received the conjugate vaccine. How- To the Editor: As a general pediatrician, I think we must ever, we find it reassuring that after several years of follow- proceed with caution when relaying the data of Eskola et up, no indication of serotype replacement has been observed al. to our patients’ parents. Although the vaccine reduced in invasive disease among the participants of the California the incidence of acute otitis media caused by pneumococcus, trial.2,3 The long-term effects of serotype replacement re- especially by the strains included in the vaccine, the overall main unknown. This warrants careful serotype-specific sur- reduction in cases of acute otitis media was only 6 percent.
veillance of the effect of vaccination on both pneumococcal The clinical significance of this finding is questionable.
carriage and invasive disease, rather than the discontinuation I have already had several parents ask me why their child of vaccination. Fortunately, data on the effect of broad-scale had an ear infection after having received the “ear-infec- vaccination on the pneumococcal serotypes causing invasive tion vaccine.” At a time when many parents are already sus- disease are rapidly accumulating in the United States, where picious of the reported safety and efficacy of vaccines, un- the pneumococcal conjugate vaccine is now in wide use.
realistic expectations may hinder the use of this vaccine andtherefore limit the documented protection it affords against 1. Black S, Shinefield H, Fireman B, et al. Efficacy, safety and immunoge-
nicity of heptavalent pneumococcal conjugate vaccine in children. Pediatr
Editor’s note: Dr. Eskola has been an employee of Aventis Pasteur since the beginning of 2000. Drs. Eskola and Kilpihave served as consultants to Wyeth Lederle Vaccines.
1. Lieu TA, Ray GT, Black SB, et al. Projected cost-effectiveness of pneu-
mococcal conjugate vaccination of healthy infants and young children.
To the Editor: It is important to make a distinction be- tween the effect of the pneumococcal conjugate vaccine on 2. Black S, Shinefield H, Fireman B, et al. Efficacy, safety and immunoge-
a larger population and the expected benefits in an individ- nicity of heptavalent pneumococcal conjugate vaccine in children. Pediatr Infect Dis J 2000;19:187-95.
ual child. Even though the proportion of vaccine-prevent- 3. Shinefield HR , Black S. The efficacy and safety of heptavalent pneumo-
able episodes of otitis media is small, the absolute number coccal conjugate vaccine in children. Presented at the Millennium Second of potentially preventable attacks at the population level is World Congress on Vaccines and Immunization, Liège, Belgium, August considerable, because otitis media is such a common dis- ease. The impact on otitis media has been estimated to ac-count for 60 percent of the cost savings expected from theprevention of pneumococcal disease through a vaccination The Effect of Risedronate on the Risk
program.1 Thus, our conclusion that the effect of the pneu- of Hip Fracture in Elderly Women
mococcal conjugate vaccine on otitis media could be sub-stantial is correct and is an important message to policy- To the Editor: The report by McClung et al. (Feb. 1 is- makers who are deciding whether the vaccine should be sue)1 is unconvincing, because information is not present- included in national immunization programs. However, we ed to show that the women in the treatment and placebo fully agree that the information given to parents should fo- groups had similar clinical risk factors for hip fracture. This cus on protection provided by the vaccine against invasive information should have been included in the table show- pneumococcal disease. Furthermore, the suggestion of tar- ing the base-line characteristics of the groups.
geting the vaccine to selected otitis-prone groups wouldbe difficult to implement and would obscure the primary objective of preventing invasive disease.
We cannot agree with Dr. Cantekin that using the num- ber of episodes instead of the number of children as exper-imental units is incorrect. The tendency to recurrence is 1. McClung MR , Geusens P, Miller PD, et al. Effect of risedronate on the
characteristic of acute otitis media. A vaccine that does not risk of hip fracture in elderly women. N Engl J Med 2001;344:333-40.
afford protection against recurrent attacks would be of littlevalue. We therefore purposely calculated the efficacy of the To the Editor: McClung et al. state that “complete fol- vaccine against all episodes of otitis media, not just the low-up data were available for 64 percent of the women.” 1720 · N Engl J Med, Vol. 344, No. 22 · May 31, 2001 · www.nejm.org
Downloaded from www.nejm.org at CILEA BIBLIOSAN on November 17, 2009 . Copyright 2001 Massachusetts Medical Society. All rights reserved. C O R R E S P O N D E N C E
In other words, follow-up data were incomplete for 36 per- various risk factors were balanced among the treatment cent, or 3324, of the 9331 women. One wonders how to groups, so the observed treatment effect was not the result interpret an estimated overall absolute difference in the inci- of differences in risk at base line. Dr. Goodman expresses dence of hip fracture of 1.1 percentage points, or 42 women appropriate concern about the potential effect of the in- with fractures, in the light of such incomplete data. A slight complete follow-up of the elderly women in our study. The preponderance of fractures among those lost to follow-up in demographic characteristics with respect to the risk of frac- the risedronate group could render the results inconclusive.
ture were very similar between the women for whom wehad complete follow-up data and those for whom we did not. Using the observed incidence of hip fractures that oc- Columbia University College of Physicians and Surgeons curred during follow-up, Dr. Goodman estimates that an additional 42 hip fractures would have been observed if allthe women had been followed for three years. If we makethe conservative assumption that there is no treatment ef- To the Editor: McClung et al. found that risedronate in fect in these women, these fractures would be proportion- addition to calcium and vitamin D reduced the risk of hip ally distributed between the treatment groups, with 14 in fracture in women 70 to 79 years old who were recruited the placebo group and 28 in the risedronate group. In this on the basis of a low bone mineral density at the femoral case, the reduction in the risk of hip fracture in the com- neck. This suggests that screening for low bone mineral den- bined risedronate groups would have remained significant sity followed by an appropriate treatment may be worth- (risk reduction, 24 percent; 95 percent confidence inter- while even in elderly women, as long as they have a very low val, 3 percent to 41 percent; P=0.03).
bone mineral density. To assess whether the incidence of The information provided by Dr. Schott and colleagues hip fracture continues to rise when the bone mineral den- from their large observational study supports our findings sity decreases to very low values, even in very elderly wom- that elderly women with fall-related risk factors do not nec- en, we reanalyzed the data from a prospective cohort study essarily have low bone density and may not be candidates for of 7598 women 75 years old or older (the Epidémiologie drug therapy to reduce the risk of hip fracture. Although de l’Ostéoporose [EPIDOS] study).1,2 The age-adjusted bone mass decreases with advancing age, we cannot assume incidence of hip fracture increased from 5.7 per 1000 per- that all older women have osteoporosis.
son-years among women with a T score of –2.5 or lower to36.5 per 1000 person-years among women with a T score lower than ¡4. In the trial of risedronate conducted byMcClung et al., no effect was observed among the women 80 years old or older. However, most women in this sub-group were enrolled on the basis of clinical risk factors only,and many did not have very low bone mineral density. In Treatment of Calcium-Channel–Blocker
the EPIDOS study, among 4478 women 80 years old or Intoxication with Insulin Infusion
older, 77 percent had one or more clinical risk factors.3 Ofthese, only 27 percent had a T score below –3. These results To the Editor: Overdose of calcium-channel blockers re- emphasize the need to measure bone mineral density at mains an important cause of fatal poisoning.1 Convention- the femoral neck to identify elderly women in whom ther- al therapy, consisting of intravenous fluids, calcium, dopa- apy to prevent fracture is appropriate.
mine, dobutamine, norepinephrine, and glucagon, oftenfails to improve hemodynamic function in intoxicated pa- tients.2 Recent recommendations for treatment of intoxi- cation with calcium-channel blockers include induction of hyperinsulinemia and euglycemia as adjunctive therapy.3 Wereport two cases of overdose of calcium-channel blockers in which a striking benefit was achieved with hyperinsuli- A 34-year-old nondiabetic woman with hypertension and renal failure ingested 12 extended-release tablets con- taining 2.5 mg of amlodipine each. Hypotension and brad- ycardia developed. Because of her history of renal failure, the treating physicians did not administer calcium and insteadprovided intravenous fluids, vasopressors, and glucagon.
1. Schott AM, Cormier C, Hans D, et al. How hip and whole-body bone
A 48-year-old nondiabetic man with hypertension, chron- mineral density predict hip fracture in elderly women: the EPIDOS Pro-spective Study. Osteoporos Int 1998;8:247-54.
ic obstructive pulmonary disease, congestive heart failure, 2. Mazess RB, Barden H. Bone density of the spine and femur in adult
and depression was witnessed ingesting an unknown white females. Calcif Tissue Int 1999;65:91-9.
amount of extended-release diltiazem. He became hemo- 3. Dargent-Molina P, Favier F, Grandjean H, et al. Fall-related factors and
dynamically unstable in the emergency department and risk of hip fracture: the EPIDOS prospective study. Lancet 1996;348:145-9. [Erratum, Lancet 1996;348:416.] did not respond to calcium, intravenous fluids, and vaso-pressors.
When conventional therapy failed to improve the hemo- dynamic status of these patients, hyperinsulinemia–eugly- To the Editor: Dr. Royce raises an important point. The cemia therapy with a continuous infusion of insulin at a rate proportions of both the older and younger women with of 0.5 IU per kilogram of body weight per hour rapidly N Engl J Med, Vol. 344, No. 22 · May 31, 2001 · www.nejm.org · 1721
Downloaded from www.nejm.org at CILEA BIBLIOSAN on November 17, 2009 . Copyright 2001 Massachusetts Medical Society. All rights reserved. The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne TABLE 1. CLINICAL COURSES OF TWO PATIENTS TREATED WITH HYPERINSULINEMIA–EUGLYCEMIA THERAPY.
TIME FROM INSULIN
INFUSION TO
VITAL SIGNS
SERUM GLUCOSE
VITAL SIGNS
RATE OF DISCONTINUATION AFTER INSULIN DURATION SUPPLEMENTAL
VITAL SIGNS
CONCENTRATION
CONVENTIONAL THERAPY
INFUSION
OF INSULIN
BEFORE THERAPY
BEFORE THERAPY
ATTEMPTED
INFUSION VASOACTIVE DRIPS
ADMINISTERED
reversed cardiovascular collapse in both. Despite the high ured by bedside capillary testing minimizes the likelihood doses of insulin administered, the first patient required no of this complication. As other case reports have suggested, supplemental glucose, whereas the second received 10 per- hyperinsulinemia–euglycemia therapy is safe and effective cent dextrose at 100 ml per hour to maintain an average for life-threatening overdose of calcium-channel blockers.3 serum glucose concentration of 140 mg per deciliter. Theclinical courses of the two patients are outlined in Table 1.
The clinical features of toxicity from calcium-channel blockers arise from blockade of L-type calcium channels inmyocardial cells, smooth-muscle cells, and beta cells.4 tagonism of these channels produces bradycardia, conduc-tion delay, peripheral vasodilation, hypoinsulinemia, hyper-glycemia, metabolic acidosis, and shock. Hypoinsulinemia 1. Litovitz TL, Klein-Schwartz W, Caravati EM, Youniss J, Crouch B, Lee
S. 1998 Annual report of the American Association of Poison Control
may be a critical factor in overdose of calcium-channel block- Centers Toxic Exposure Surveillance System. Am J Emerg Med 1999;17: ers.5 In an unstressed state, myocytes oxidize free fatty acids for metabolic energy.3,4 In a state of shock, such as that as- 2. Enyeart JJ, Price WA, Hoffman DA, Woods L. Profound hyperglycemia
sociated with overdose of calcium-channel blockers, myo- and metabolic acidosis after verapamil overdose. J Am Coll Cardiol 1983;2:1228-31.
cytes use glucose for fuel.3,4 Hypoinsulinemia may prevent 3. Yuan TH, Kerns WP II, Tomaszewski CA, Ford MO, Kline JA. Insulin-
the uptake of glucose by myocytes, causing a loss of inotro- glucose as adjunctive therapy for severe calcium channel antagonist poison- py, decreased peripheral vascular resistance, and shock.3 ing. J Toxicol Clin Toxicol 1999;37:463-74.
The exact mechanism of action of hyperinsulinemia– 4. Kline JA, Leonova E, Raymond RM. Beneficial myocardial metabolic
effects of insulin during verapamil toxicity in the anesthetized canine. Crit
euglycemia therapy is poorly defined. Hyperinsulinemia– euglycemia therapy improves inotropy and peripheral vas- 5. Ohta M, Nelson J, Nelson D, Meglasson MD, Erecinska M. Effect of
cular resistance and reverses acidosis, possibly by improving Ca++ channel blockers on energy level and stimulated insulin secretion in the uptake of carbohydrates by myocytes and smooth-mus- isolated rat islets of Langerhans. J Pharmacol Exp Ther 1993;264:35-40.
cle cells.3,4 Although hypoglycemia may occur with thistherapy, the ease with which serum glucose can be meas- Correspondence Copyright 2001 Massachusetts Medical Society.
1722 · N Engl J Med, Vol. 344, No. 22 · May 31, 2001 · www.nejm.org
Downloaded from www.nejm.org at CILEA BIBLIOSAN on November 17, 2009 . Copyright 2001 Massachusetts Medical Society. All rights reserved.

Source: http://www.galloniprosciutto.it/files/allegati/na%20and%20dietary%20letter.pdf