Practical Pointers

1) The Highlights section contains brief comments patterned after the “abstract” placed on the first page of many studies reported in journals. Highlights condenses the content of studies, and allows a quick review of pertinent points of each article.

The Editorial Comments are the editor’s assessments of the clinical practicality of articles based on his long-term review of the current literature and his 20-year publication of Practical Pointers.

An Index containing all the Highlights is published twice a year. In an evening or two, the reader can refresh memory of the entire content of practical points abstracted from 6 major journals.

2) The main Abstracts section is designed as a reference. It presents structured summaries of the content of articles in much more detail.

I hope you will find Practical Pointers interesting and helpful. The complete content of all issues for the past 5 years can be accessed at www.practicalpointers.org

Richard T. James Jr, M.D.

Editor/Publisher.

HIGHLIGHTS AND EDITORIAL COMMENTS JUNE 2005

Reduced Incidence And Severity Of HZ And PHN.

6-1 A VACCINE TO PREVENT HERPES ZOSTER AND POSTHERPETIC NEURALGIA IN OLDER ADULTS.

This study tested the hypothesis that vaccination would decrease the incidence and severity of both HZ and PHN.

Over 38 000 subjects were randomized to: 1) a subcutaneous injection of live, attenuated varicella-zoster vaccine, or 2) placebo. The potency of the live attenuated Oka vaccine was about 14 times that of the varicella vaccine given to children.

A. Herpes zoster: (3 years) Vaccinated Placebo Absolute difference NNT

Confirmed cases of acute HZ 315 642 1.7% 58

Overall incidence of HZ

per 100 person-years 0.54 1.11 0.57% 175

Median duration of pain (days) 21 24

Severity of illness 141 180 (area under the curve)

Burden of illness score 2.2 5.7

B. Postherpetic neuralgia (3 years)

Confirmed cases 27 80 0.3% 333

Persistence of pain was shorter in the vaccinated group.

There was no evidence that the live vaccine caused HZ.

Adverse effects were generally mild, mainly due to local reactions.

We would expect more cases of HZ would be prevented as time progressed, and as more individuals enter the ranks of the elderly. I asked myself—at my advanced age should I take the vaccine? Having seen the devastating complications of zoster, I would be more than willing to take it, even though the likelihood of prevention of HZ over 3 years is only 1 in 58. .

Questions remain:

At what age should the vaccine be recommended?

How long is the boost in immunity protective?

Is it cost-effective enough for Medicare to cover costs?

Antibiotics Provided Little Advantage Compared With No-Antibiotics.

6-2 INFORMATION LEAFLET AND ANTIBIOTIC PRESCRIBING STRATEGIES FOR ACUTE LOWER RESPIRATORYINFECTION

Pharyngitis and acute bronchitis are the main causes of excess antibiotic prescribing.

This pragmatic study assessed the effectiveness of 3 different antibiotic strategies for acute bronchitis.

Randomized, controlled trial followed over 800 patients presenting to primary care with acute uncomplicated LRI. Patients with findings suggestive of pneumonia were excluded—new focal chest signs (focal crepitations or bronchial breathing); and systemic features (high fever, vomiting, severe diarrhea). Also excluded patients with asthma, other chronic or acute lung diseases, cardiovascular disease, or with previous pneumonia.

Randomized to: 1) no antibiotic prescribed [control group], 2) delayed prescription [to be picked up later], or 3) immediately prescribed antibiotic. The antibiotic of choice was amoxicillin 250 mg 3 times daily for 10 days, or, if allergic, erythromycin 250 mg 4 times a day for 10 days.

Compared with no antibiotics [control group], the other strategies did not significantly alter cough duration: Delayed prescription shortened duration by 0.75 days; immediate prescription by 0.11 days. Treatment group had no effect on duration of other symptoms.

“Compared with immediate antibiotics, a strategy of either no offer of antibiotics or a delayed prescription was associated with little difference in duration or severity of symptoms.” Overall, antibiotics probably do provide modest symptomatic relief. If a benefit is present, it represents a shortening of only one day in a relatively long history. “It is difficult to justify widespread antibiotic prescribing for uncomplicated lower respiratory infection on this basis, given the dangers of antibiotic resistance.”

I was somewhat surprised at the duration of cough symptoms in this group of patients—a mean total of 3 weeks. However, I believe most patients would experience a gradual improvement over this period. We are admonished to consider pertussis in patient with LRI when the cough lasts 3 weeks or more. I presume in pertussis the cough continues unabated.

I believe advising patients that antibiotics may be associated with serious adverse effects (eg, colitis) will do more to tilt them toward accepting only symptomatic therapy than would advising them of the danger of antibiotic resistance in the community.

I have had success in prescribing delayed prescriptions of patients with uncomplicated lower respiratory infections. The great majority never fills the prescription. This may be an acceptable means of satisfying a demanding patient.

In the US, It is likely that many patients presenting after a week or more of cough and sputum production will receive a cheat X-ray.

The decision by primary care clinicians to prescribe or not prescribe, I believe, will often depend on how “sick” the patient appears.

A Circumference Under 100 Cm Rules Out Insulin Resistance And Hyper-Insulinemia.

6-3 USE OF WAIST CIRCUMFERENCE TO PREDICT INSULIN RESISTANCE

This study assessed how effectively different anthropometric markers used in clinical practice can predict insulin sensitivity. The authors suggest abdominal circumference is the most powerful independent predictor to rule out insulin resistance.

Determined height, weight, and waist circumference (midway between lateral lower ribs and iliac crest). Also determined results from analyses of plasma for glucose, insulin, and lipid concentrations. Used a homoeostasis index as a measure of insulin sensitivity [plasma glucose (mol/L) X plasma insulin (mU/L)/22.5]. A score of 4.0 and greater was defined as insulin resistance.

Using 100 cm as a test, the authors determined the sensitivity to diagnose insulin resistance was between 94-98%; and the specificity was between 61-63%. The positive predictive values of the test were 61% for men and 42% for women. The negative predictive value of the test was 98% in men and 97% in women. A waist circumference under 100 cm was therefore a strong independent predictor in ruling out insulin resistance.

Waist circumference is a simple tool to exclude insulin resistance. If the patient has a circumference under 100 cm (40 inches), he or she is very unlikely to have insulin resistance and hyper-insulinemia. Circumferences above 100 cm may, or may not, be related to insulin resistance.

I found this short article provocative. The results require confirmation. Abdominal girth is an important risk factor for the metabolic syndrome and cardiovascular disease.

6-4 SENSITIVITY, SPECIFICITY, AND PREDICTIVE VALUES

The authors of the preceding article calculated the sensitivity, specificity, and predictive values of the waist circumference test.—the ability of the test to detect insulin resistance and insulin sensitivity among healthy subjects by using 100 cm as a cut-off point. I welcome opportunities to review these important statistical measures. I have done so many times. I still have some difficulty in thinking them through and calculating them accurately. I used the determinations in the article as an example.

See the abstract.

“No Indication Of A Net Benefit.”

6-5 EPIDEMIOLOGICAL MODELLING OF ROUTINE USE OF LOW-DOSE ASPIRIN FOR THE PRIMARY PREVENTION OF CORONARY HEART DISEASE AND STROKE IN THOSE AGE > 70

Current US guidelines recommend the use of low-dose aspirin for people with a 5-year absolute risk of coronary heart disease (CH D) of > 3%, or a 10-year absolute risk of > 10%.

“Prophylactic use of a potentially toxic agent can be problematic, particularly in people in whom comorbidity and polypharmacy are common.” In a prospective observational study in two large UK general hospitals, aspirin use was the causal agent in 18% of all admissions for adverse drug effects, and was implicated in 61% of all associated deaths. Older females are the most vulnerable.

This epidemiological modeling study was conducted in a hypothetical population (10 000 men and 10 000 women) selected from a reference population from a state in Australia. All were age 70-74. None had known cardiovascular disease.

Proportional benefit gained from aspirin in prevention of MI and ischemic stroke vs excess hemorrhage from age 70-74 to age 100 or to death:

Benefit in preventing Men (n = 10 000) Women (n = 10 000)

Myocardial infarction - 389 - 321

Ischemic stroke - 19 - 35

Harm

Excess GI hemorrhage + 499 + 572

Excess hemorrhagic stroke + 76 + 54

When comparing net harms vs net benefits of aspirin, the effects on length and quality of life were equivocal.

“Despite sound evidence for efficacy, the temptation to blindly implement low-dose aspirin treatment for the primary prevention of cardiovascular disease in elderly people must be resisted.” Benefits may be offset by harms.

I believe low-dose aspirin has an important place in primary prevention of women at higher risk, and in secondary prevention of cardiovascular disease.

There is an important clinical downside related to universal prophylactic aspirin therapy: suppose primary care clinicians prescribe low-dose aspirin to 1000 women over 10 years. Three or 4 ischemic strokes might be prevented. But there would be no way of knowing which individuals of the 1000 benefited. Conversely, a serious hemorrhagic event occurring in 2 of the1000 patients would be self-evident. The clinician might feel responsible, and the patient and family might blame the clinician for the disaster.

I believe primary prevention with aspirin in women at average risk should be avoided. Obviously, careful clinical judgment based on individual-patient attributes is required.

“Our Most Important Finding Was The High Level Of Antipsychotic Prescribing In NHs.”

6-6 THE QUALITY OF ANTIPSYCHOTIC DRUG PRESCRIBING IN NURSING HOMES

Antipsychotic drug prescribing in nursing homes (NHs) has been rising.

Federal statutes are in effect to protect NH residents from receiving inappropriate antipsychotics. They may be appropriately prescribed for delirium and dementia only if psychotic features or dangerous behaviors are present. Guidelines also stipulate maximum daily doses.

For residents with dementia, behavioral assessments must also show evidence of verbal or physical aggression or delusions or hallucinations.

Impaired memory, wandering, restlessness, unsociability, uncooperativeness, and indifference to surroundings are NOT indications.

Use of antipsychotic drugs in NHs was widespread. Most atypicals were prescribed outside the prescribing guidelines with doses, and for indications without strong clinical evidence of benefit. About 1 in 4 received doses exceeding recommended. About 2/3 of use was appropriate—dementia with aggressive behavior; dementia with delusions; psychotic disorder. About 1/3 received the drugs inappropriately—impaired memory; depression without psychotic features; indifference to surroundings; insomnia; anxiety; wandering; restlessness; uncooperativeness; unsociability.

The study failed to detect positive relationships between behavioral symptoms and antipsychotic therapy.

“This study raises questions about the current uses of antipsychotics in NHs.”

These are powerful drugs. Elderly patients are subject to more adverse effects. They require a lower dose because of impaired renal function and concomitant illness. The PDR reiterates that schizophrenia is the only indication. There is no mention of use in nursing homes. Few studies have concerned patients over age 65.

I believe the most appropriate question to ask when contemplating use of antipsychotics in NHs is…Am I prescribing this drug to benefit the patient, or the nursing staff and the family? This can be a most difficult decision to make. If they are prescribed, individual- patient’s response must be carefully monitored.

When To Intervene? How To Intervene?

6-7 Thresholds For Normal Blood Pressure And Serum Cholesterol.

In 2003, European guidelines suggested a BP of above 140/90, and a cholesterol above 5 mmol/L (193 mg/dL) as the appropriate thresholds for intervention. “The bottom line is that the doctor is expected to inform the patient that these measurements mean that he or she is at increased cardiovascular risk regardless of the management proposed. In other words, a disease label is to be attached to the patient.”

In Norway, if this threshold for cholesterol and BP were to be applied at age 24, half the population would be identified as being at increased risk. At age 49, the proportion is raised to 90%. As much as 75% of the total population would be identified as being at risk.

The potential benefits for treated patients become less at lower risk levels. The number needed to treat is increased. The rates of adverse effects (of drug treatment) remain the same. Adverse effects tend to be under-reported and under-published.

Certainly, experts who developed these guidelines did not suggest that all persons with BP and cholesterol levels above these cut-points should be treated with drugs.

I believe however, that all should be treated with judicious advice about changing in lifestyle. This will apply to almost all persons in the US over age 50. Very rarely will individuals over age 50 have no risk factors. Cut-points are defined at levels below which no further reduction in risk occurs. Admittedly, those with baseline risk-levels at the low range will have less to gain when their levels are lowered than persons with high baseline risk-levels.

I do not believe life-style advice will be interpreted as a labeling of disease. There are few if any adverse effects of lifestyle changes. Effectiveness is established. The benefit/harm-cost ratio is very low.

The task of educating patients about healthy lifestyles and getting them to adopt them is daunting, and in the main unsuccessful. We should not be deterred from trying. This includes primary care clinicians’ adopting a healthy lifestyle themselves.

Who should be treated with drugs?—patients who are indeed at high risk. The definition of “high risk” depends not only on the number or risk factors present and their levels, but also on the individual patient’s assessment of his own risk. Patients must be convinced of the benefits of drug therapy; must understand that drug therapy is long-term, expensive, and carries risks of its own.

No Evidence Of Superiority For Treatment With A Calcium Channel-Blocker, Or An ACE Inhibitor Compared With A Thiazide-Type Diuretic

6-8 CLINICAL OUTCOMES IN ANTIHYPERTENSIVE TREATMENT OF TYPE 2 DIABETES, IMPAIRED FASTING GLUCOSE CONCENTRATION, AND NORMOGLYCEMIA

The Antihypertensive and Lipid-lowering Treatment to prevent Heart Attack Trial (ALLHAT)

This is one of a series of articles reported by the ALLHAT group.

There was no evidence of superiority for treatment with a calcium channel-blocker, or an ACE inhibitor compared with a thiazide-type diuretic during first-step antihypertension therapy in DM, IFG, or NG.

Most hypertensive patients with DM or IFG or impaired glucose tolerance, I believe, would receive more than one antihypertension drug. Many clinicians would use a combination of an ACE inhibitor and a diuretic.. Any combination should include a thiazide.

The abstract of this study is brief since I already abstracted similar studies by the same ALLHAT group: A. JAMA December 18, 2002; 2981-97 presented the original ALLHAT study. (See Practical Pointers December 2002 [12-1] ) The study compared the same 3 drugs in similar patients with hypertension and at least one additional risk factor (high-risk) . Conclusion: “Thiazide-type diuretics should be considered first for pharmacological therapy in patients with hypertension. ” They are unsurpassed in lowering BP, reducing clinical events, and in tolerability. They are much less costly. Since many patients with hypertension will require more than one drug to control their BP, it is reasonable to infer that a diuretic should be included in all multidrug regimens.

B. JAMA April 6, 2005; 293: 1595-1608 (See Practical Pointers April 2005 [4-2] ) “Thiazide-type diuretics remain the drugs of first choice for initial therapy of hypertension in both black and non-black hypertensive patients.”

C. Archives Int Med April 25, 2005; 165: 936-46. “Renal Outcomes in High-Risk Hypertensive Patients with an Angiotensin-Converting Enzyme Inhibitor or a Calcium Channel Blocker vs Diuretic.” In hypertensive patients with reduced glomerular filtration rate, neither amlodipine nor lisinopril was superior to chlorthalidone in reducing the rate of development of end-stage renal disease or a 50% or greater reduction of glomerular filtration rate.

See also:

NEJM December 30, 2004; 351: 2805-16 (Practical Pointers December 2004 [12-2] ) “Association between Cardiovascular Outcomes and Antihypertensive Treatment in Older Women” Conclusion: Monotherapy with diuretics was equally or more effective than other monotherapies. The combination of diuretics + beta-blockers was superior to, or equally effective as, other combinations.

Likely To Produce Clinically Important Pain Relief.

6-9 EFFICACY AND SAFETY OF OPIOID AGONISTS IN THE TREATMENT OF NEUROPATHIC PAIN OF NON-MALIGNANT ORIGIN A Systematic Review.

Effective pain relief in these patients is difficult to achieve. Use of opioids is controversial. This is in part because studies have been small, have yielded equivocal results, and have not established long-term efficacy and safety. There have been concerns about adverse effects: potential for abuse and addiction, hormonal abnormalities, dysfunction of the immune system, and paradoxical hyperalgesia.

This systematic review assessed the efficacy and safety of opioids for treatment of NP.

Opioid treatment for 1 to 8 weeks had a beneficial effect over placebo for spontaneous neuropathic pain. The magnitude of this opioid effect was nearly a 14-point difference in pain intensity at study end compared with placebo. Is an average decline of 14 points on a 100-point scale meaningful to patients? The mean initial pain intensity ranged from 46 to 69. A 14-point difference corresponds to a 20% to 30% reduction in pain.

The trial did not address issues of addiction. It is reasonable to assume that the studies did not include individuals who might be potential abusers.

The most common adverse effects were nausea, constipation, drowsiness, vomiting, and dizziness.

(NNT to harm one patient = 4 to 7.)

I believe primary care clinicians underuse opioids for patients with non-cancer pain. Fears of addiction have been overemphasized.

Donepezil May Delay Clinical Progression To Alzheimer’s Disease

6-10 VITAMIN E AND DONEPEZIL (ARICEPT) FOR THE TREATMENT OF MILD COGNITIVE IMPAIRMENT

Amnestic (memory loss) mild cognitive impairment (MCI) represents a transitional state between the cognitive changes of normal aging and the earliest clinical features of Alzheimer’s disease (AD). Amnestic MCI refers to the subtype that has a primary memory component, either alone or in conjunction with other cognitive-domain impairments, of insufficient severity to constitute dementia. About 80% of those who meet the criteria for MCI will have AD within 6 years.

MCI is a transition state between normal aging and dementia (for Alzheimer’s disease in particular), one in which cognitive deficits are present, but function preserved. In clinical settings, the term is often used to describe patients who present with memory loss, but do not have dementia. Even when defined carefully, MCI is a heterogeneous category that includes some persons with memory changes of normal aging, some with non-progressive cognitive defects, some with prodromal AD, and some with prodromal forms of other neurodegenerative dementias.

This study was designed to determine if vitamin E or the cholinesterase inhibitor donepezil could delay the clinical diagnosis of AD in patients with MCI.

Vitamin E had no effect at any time.

For donepezil . . . “The observed relative reduction in the risk of progression of 56% at one year and 36% at two years in the entire cohort is likely to be clinically significant.”

“Although our findings do not provide support for a clear recommendation for the use of donepezil in persons with mild cognitive impairment, they could prompt a discussion between the clinician and the patient about this possibility.”

Symptoms of memory loss in older persons should be taken seriously. They may represent the beginning of AD. This may be an important clinical measure once more effective treatments become available.

The important question is . . . What are the cognitive changes of normal aging?

I believe some degree of memory impairment is almost universal among individuals over age 80. It usually begins by forgetting names, and recalling them minutes or hours later ( “senior moments”). The spectrum of memory impairment is wide. The definition of amnestic MCI is not settled. At what point does it predict development of AD? The criteria for diagnosis of amnestic MCI in the study included patients with difficulties greater than temporarily forgetting names.

This study may foretell important developments in drug therapy which may delay the onset of disabling dementia. The spectrum of forgetfulness of old age is very broad. When should intervention be considered? Some elderly patients may well accept early intervention. A delay of one to two years represents a large proportion of remaining quality-life. Patients may be willing to accept some adverse effects of drugs to gain a few years free of dementia of AD. (Note that anticholinergics do not benefit vascular dementia.)

Others may wish to wait until adverse effects on daily living become more evident.

I do not believe memory defects inevitably progress to AD. Keeping mentally and physically active, continuing a healthy diet, retaining active family and social connections, and controlling risk factors for cardiovascular disease will delay or prevent development of dementia in many individuals.

In Modestly Overweight Persons, Reduction In Weight May Lower Risk Of Developing Hypertension.

6-11 WEIGHT LOSS IN OVERWEIGHT ADULTS AND THE LONG-TERM RISK OF HYPERTENSION: The Framingham Study

Obesity is associated with higher levels of insulin resistance, hyperinsulinemia, rises in cardiac output, increases in cholesterol and triglycerides, and increased sympathetic nervous system activity. Most of these changes have been associated with increases in BP. “In recent years, there has been a great deal of focus on the roles of hyperinsulinemia and insulin resistance in the development of hypertension.”

The goal of this study was to estimate the effects of both the amount on weight loss and the persistence of weight loss on the risk of incident hypertension among already obese adults. (Primary prevention.)

After multiple adjustments, weight loss of 6.8 kg (18 pounds) or more led to a 28% reduction in risk of developing long-term hypertension in younger subjects (mean age 27) , and a 37% reduction in older subjects (mean age 52).

If the weight loss was sustained over the years, the risks of developing hypertension were reduced by 22% and 26%.

“The results of this study suggest that at least 15% of the cases of hypertension in overweight middle-aged adults and 22% of the cases occurring in overweight older adults could be prevented by a modest amount of sustained weight loss.”

Overweight + hyperinsulinism + dyslipidemia + hypertension = a common and deadly combination

Treatment Reduced Serious Perinatal Morbidity In Infants And May Improve The Woman’s Health-Related Quality Of Life.

6-12 GESTATIONAL DIABETES MELLITUS; Effect Of Treatment On Pregnancy Outcomes.

Gestational diabetes mellitus (GDM) occurs in up to 9% of all pregnancies. It is associated with substantial maternal and perinatal complications. Neonatal complications include macrosomia, shoulder dystocia, birth injuries, bone fractures, nerve palsies, and hypoglycemia. Long-term adverse health outcomes among infants born to mothers with GDM include sustained glucose intolerance, subsequent obesity, and impaired intellectual achievement.

This study asked . . . Does screening and treatment for GDM reduce these risks?

This randomized trial enrolled 1000 women who were between 16 and 30 weeks pregnant. Randomized to: 1) An intervention group received expert diabetes care including education, self-monitoring blood glucose, and adjusted insulin therapy, and 2) A usual care group.

Serious perinatal complications in infants were significantly lower in the intervention group (1% vs 4%). The NNT to prevent one serious outcome in infants = 34. Birth weights were lower in the intervention group (less likely to have macrosomia). No difference in rate of hypoglycemia requiring intravenous glucose.

Women in the intervention group gained less weight and had less risk for preeclampsia. At 3-months postpartum, women had lower rates of depression and higher scores on quality-of-life. Rates of caesarean deliveries were similar.

Impaired glucose tolerance and diabetes are important risk factors at the time of conception. Primary care clinicians can serve their young adult female patients by advising them of the risks of glucose intolerance (and excessive weight) before and at the time of conception.

The Test Could Be Used As An Initial Approach To Diagnosis.

6-13 IS PROTON PUMP INHIBITOR TESTING AN EFFECTIVE APPROACH TO DIAGNOSE GASTRO ESOPHAGEAL REFLUX DISEASE IN PATIENTS WITH NON-CARDIAC PAIN? A Meta-analysis

Gastro esophageal reflux disease (GERD) is the most common cause of non-cardiac chest pain (NCCP). Patients with NCCP are often treated empirically and successfully with proton pump inhibitors.

This study asked. . .Can proton pump inhibitors ( a PPI test) be used as a diagnostic test?

Results of the PPI test:

GERD present GERD absent

Positive test (> 50% relief) 80% (true positive)* 26% (false positive)

Negative test (< 50% relief) 20% (false negative) 74% (true negative test)**

(* sensitivity of the PPI test = true + % = 80%; ** specificity of the PPI test = true negative % = 74%)

Results of the placebo test:

Positive test (> 50% relief) 19% (true positive)*** 23% (false positive)

Negative test (< 50% relief) 81% (false negative) 77% (true negative test)****

(*** sensitivity of placebo test =19%; **** specificity of placebo test = 77%)

Thus 80% responded favorably to PPI vs 19% to placebo.

Treatment with PPIs and placebo showed similar effects (26% and 23%) on improving NCCP symptoms

in patients without GERD, indicating a possible placebo effect.

The use of PPI as a diagnostic test for detecting GERD in patients with NCCP has an “acceptable” sensitivity and specificity and could be used as an initial approach by primary care physicians to detect GERD in selected patients with NCCP.

“Acceptability of the test would be more meaningfully determined by calculating pre-test probability., likelihood ratios, and post-test probability. See the full abstract.

Regardless of the modest diagnostic help given by a PPI test, I believe, in practice, the test is used extensively by primary care clinicians and their patients.

May Reduce Risk Of Development Of PMS.

6-14 CALCIUM AND VITAMIN D INTAKE AND RISK OF INCIDENT PREMENSTRUAL SYNDROME

Several studies have suggested that calcium and vitamin D levels are lower in women with PMS, and that calcium supplementation may prevent the initial development of PMS.

This case-control study was nested within the large prospective Nurses’ Health Study. Participants were a subset of women age 27 to 44 (mean = 35). All were free of PMS at baseline (1991). Cases: 1057 women who developed PMS over a 10-year follow-up. Controls: 1968 women who reported no diagnosis of PMS and no, or minimal, menstrual symptoms.

Determined dietary and supplemental intakes of calcium and vitamin D by periodic questionnaires.

Women in the highest quintile of total vitamin D intake (median of 706 IU) had a relative risk of new-onset PMS of 0.59 compared with those in the lowest quintile (median of 112 IU). Benefit was associated with vitamin D from food. Supplemental vitamin D did not seem to be associated with risk.

Similar benefit was associated with calcium intake from food.

I abstracted this article because its conclusions are provocative—certainly not definitive. It raises more questions: Why did the benefit not extend to supplements? Is there a reasonable biological mechanism for the action of calcium and vitamin D? Why no benefit from whole milk? At a more practical level—could diet be beneficial in treatment as well as prevention?

I will watch for more developments.

Hope For Reducing The Prevalence Of This World-Wide Scourge.

6-15 BASIC SCIENCE GUIDELINES DESIGN OF NEW TB VACCINE CANDIDATES

Several new vaccines which improve immune response are under investigation. They may be helpful in primary prevention of infection as well as boosting immunity in those with latent infection.

See the abstract.

(The entire June 8 2005 issue of JAMA is devoted to tuberculosis.)

ABSTRACTS JUNE 2005

Reduced Incidence And Severity Of HZ And PHN.

6-1 A VACCINE TO PREVENT HERPES ZOSTER AND POSTHERPETIC NEURALGIA IN OLDER ADULTS.

Cell-mediated immunity declines with age. This increases likelihood of reactivation of the latent varicella-zoster virus within the sensory ganglia, causing herpes zoster (HZ). Complications of HZ occur in about 50% of elderly patients, most commonly postherpetic neuralgia (PHN).

The pain of PHN can be prolonged and disabling. It can diminish the patient’s quality of life and ability to function to a degree comparable to that of diseases such as congestive heart failure, myocardial infarction, type 2 diabetes, and major depression.

Antiviral therapy reduces the severity and duration of the acute phase of HZ, but does not prevent PHN.

Recurrences of HZ are uncommon among immunocompetent persons because an episode of acute HZ boosts immunity, effectively “immunizing” against a subsequent attack.

Vaccines can elicit a significant increase in cell-mediated immunity to HZ in immunocompetent older adults.

This study tested the hypothesis that vaccination would decrease the incidence and severity of both HZ and PHN.

Conclusion: The vaccine markedly reduced incidence and severity of HZ and PHN.

STUDY

1. A randomized, double-blind placebo-controlled trial enrolled over 38 000 adults. All were over age 60 (46% over age 70).

2. Randomized to: 1) a subcutaneous injection of live, attenuated varicella-zoster vaccine, or 2) placebo. The potency of the live attenuated Oka vaccine was about 14 times that of the varicella vaccine given to children.

3. Determined incidence, severity and duration of subsequent acute HZ, and incidence of PHN.

RESULTS

1. Over 95% of subjects completed the 3-year study.

2. Outcome:

A. Herpes zoster: (3y) Vaccinated Placebo Absolute difference NNT

Confirmed cases of acute HZ 315 642 1.7% 58

Overall incidence of HZ

per 100 person-years 0.54 1.11 0.57% 175

Median duration of pain (d) 21 24

Severity of illness 141 180 (area under the curve)

Burden of illness score 2.2 5.7

B. Postherpetic neuralgia (3y)

Confirmed cases 27 80 0.3% 333

Persistence of pain was shorter in the vaccinated group.

3. During the first 42 days after vaccination, 7 cases of HZ occurred in the vaccine group vs 24 in the placebo group. (Apparently the boost in immunity occurs quickly.)

4. The results of PCR testing were positive for wild-type varicella-zoster virus DNA in over 93% of cases. The vaccine virus DNA was not detected in any case. (Ie, the live vaccine does not cause HZ.)

6. Adverse events:

Serious adverse events—number and types of were similar in both groups.

Local reactions (especially erythema and tenderness) were more common in the vaccine group. They were usually mild.

Hospitalizations were similar and not considered related to the vaccine. No vaccine recipient developed fever over 38.3⁰

DISCUSSION

1. An estimated 1 million cases of HZ occur in the US each year. The number is expected to increase as the population ages.

2. In older subjects in the study, the benefit of the vaccine in reducing incidence of HZ was less, but the effect on reducing severity of the illness was greater.

3. The potency of the vaccine was much greater than the vaccine given to children. “We do not recommend the use of the current varicella vaccine in an attempt to protect against herpes zoster and postherpetic neuralgia.”

CONCLUSION

The vaccine markedly reduced morbidity from acute HZ, and PHN.

NEJM June 2, 2005; 352: 2271-84 Original investigation by the Shingles Prevention Study Group, first author M N Oxman, VA San Diego Health-care System, CA.

An editorial in this issue of NEJM (pp 2344-46) by Donald H Gilden, comments:

Should vaccine be recommended for all VZV-seropositive middle-aged adults who have not had zoster? The author suggests that grown-ups should welcome the vaccine. “We may need it more than children do.” He cites 2 considerations:

1) What is the future risk of zoster among adults who were vaccinated for chickenpox in childhood? At present, most elderly people have not been vaccinated. By 2047, most middle-aged Americans will have received chickenpox vaccine. They have not had chickenpox. Like the wild-type virus, the live Oka varicella vaccine virus given to children becomes latent in ganglia. If the viral burden is less in adults who were vaccinated in infancy—compared with adults who have had chickenpox in childhood—then the incidence of zoster may be reduced. Would the vaccine be indicated in this group?

2) Cost-effectiveness: This depends on price. It may be that the adult vaccine, given its greater potency, will be more expensive than the childhood vaccine. Nevertheless, the zoster vaccine appears to be highly cost-effective even assuming a cost of $500.

================================================================================

“No Difference in Symptom Relief”

6-2 INFORMATION LEAFLET AND ANTIBIOTIC PRESCRIBING STRATEGIES FOR ACUTE LOWER RESPIRATORYINFECTION

Pharyngitis and acute bronchitis are the main causes of excess antibiotic prescribing. Costs each year exceed $700 million. A consensus has been made for limiting use of antibiotics in lower respiratory infections. (LRI). But recent systematic reviews have come to diverse conclusions about the effectiveness of antibiotics. The most recent Cochrane review confirms a moderate benefit of antibiotics on the course of the illness, but called for more evidence. Reviews have concerned relatively small numbers of patients.

Strategies to treat upper respiratory which do not include initial antibiotics—either no antibiotics or delayed antibiotics—are effective in up to 90% of cases, result in acceptable symptom control, are satisfactory to the patient, and can reduce reconsultation by up to 40%.

The debate about use of antibiotics for lower respiratory infections continues.

This pragmatic study assessed the effectiveness of 3 different antibiotic strategies, and the effectiveness of an information leaflet given to patients compared with brief verbal information alone.

Conclusion: There was no difference in symptom resolution between immediate prescription of antibiotic, no prescription for antibiotic, and delayed prescription. The information leaflet was of little help.

STUDY

1. Randomized, controlled trial followed over 800 patients presenting to primary care with acute uncomplicated LRI. Patients with findings suggestive of pneumonia were excluded—new focal chest signs (focal crepitations or bronchial breathing); and systemic features (high fever, vomiting, severe diarrhea). Also excluded patients with asthma, other chronic or acute lung diseases, cardiovascular disease, or with previous pneumonia.

2. The patients included 17% under age 16 and 17% over age 60. (mean age 39). They were moderately ill. The majority had fever, sore throat, and coryza. Many were producing dark green sputum. Some had coarse crepetitations and wheeze.

3. The mean duration of cough before the first consultation was 9 days.

4. Randomized to: 1) no antibiotic prescribed [control group], 2) delayed prescription [to be picked up later], or 3) immediately prescribed antibiotic. The antibiotic of choice was amoxicillin 250 mg 3 times daily for 10 days, or, if allergic, erythromycin 250 mg 4 times a day for 10 days.

5. The delayed prescription was left in a box in the reception room of the office to be picked up at any time the patients wished.

6. Patients were also randomized to receive, or not receive, an information leaflet describing the natural history of acute LRI. (I omit this data, since it provided little benefit compared with a brief oral consultation. RTJ)

7. Patients were asked to record their temperatures. They also were asked to keep a diary and record use of antipyretics, and severity of 6 symptoms—cough, dyspnea, sputum production, well-being, sleep disturbance, and activity disturbance.

RESULTS

1. After the consultation, cough rated as at least a “slight problem” lasted for a mean of 12 days (25% over 17 days).

2. Compared with no antibiotics [control group], the other strategies did not significantly alter cough duration: delayed prescription shortened duration by 0.75 days; immediate prescription by 0.11 days. Treatment group had no effect on duration of other symptoms.

3. Immediate Delayed Control (none prescribed)

Use of antibiotics: 96%; 20%; 16%.

Patients “very satisfied” 86%; 77%, 72%.

Believed antibiotics to be effective 75% 40% 47%

Re-attendance within 1 month 11% 12% 19%

(Note that relatively few of the delayed group actually had prescriptions filled.)

4. Adverse events; one patient (of 212) in the no antibiotic group developed pneumonia. He was hospitalized and treated with antibiotics. He recovered fully.

DISCUSION

1. There is no widely agreed definition of lower respiratory infection. In practice, most patients have cough with sputum.

2. “Our study confirms the long history of lower respiratory infection. Patients need to be warned that they will on average have an illness lasting 3 weeks in total with10 days of symptoms before the physician visit and 12 days after the physician visit.”

3. “Compared with immediate antibiotics, a strategy of either no offer of antibiotics or a delayed prescription was associated with little difference in duration or severity of symptoms.” Overall, antibiotics probably do provide modest symptomatic relief. If a benefit is present, it represents a shortening of only one day in a relatively long history. “It is difficult to justify widespread antibiotic prescribing for uncomplicated lower respiratory infection on this basis, given the dangers of antibiotic resistance.”

4. A high percentage of patients receiving no antibiotics or delayed prescription for antibiotics were satisfied with their treatment.

5. The practice of no offer of antibiotics and delayed prescription for antibiotics is likely to reduce use of antibiotics.

CONCLUSION

For patients with LRI, not prescribing antibiotics or offering a delayed prescription (compared with immediate prescription) was associated with little difference in symptom resolution.

JAMA June 22/29, 2005; 293: 3029-35 Original investigation, first author Paul Little, University of Southampton, UK.

A Circumference Under 100 Cm Rules Out Insulin Resistance And Hyper-Insulinemia.

6-3 USE OF WAIST CIRCUMFERENCE TO PREDICT INSULIN RESISTANCE

Insulin resistance is an important component of the metabolic syndrome. No easy clinical test exists to determine insulin resistance in an individual.

This study assessed how effectively different anthropometric markers used in clinical practice can predict insulin sensitivity. The investigators analyzed a sample of 2746 healthy volunteers (798 male) from retrospectively collected data. Ages ranged from 18 to 72; body mass indexes from 18 to 60; and waist circumferences from 65 cm to 150 cm.

Assessed predictive power of 5 variables: waist circumference; plasma triglycerides; systolic BP; HDL-cholesterol; and body mass index.

Set the optimal abdominal circumference cut-point for detecting insulin resistance at 100 cm (40 inches) for both men and women. Using 100 cm as a test, the authors determined the sensitivity to diagnose insulin resistance was between 94-98%; and the specificity was between 61-63%. The positive predictive values of the test were 61% for men and 42% for women. The negative predictive value of the test was 98% in men and 97% in women. A waist circumference under 100 cm was therefore very predictive in ruling out insulin resistance. (These figures depend on the prevalence of insulin resistance in the actual sample.)

COMMENT

1. Some studies report that the prevalence of the metabolic syndrome is similar in both sexes.

2. “A waist circumference under 100 excluded individuals of both sexes from the risk of being insulin resistant.”

3. At a cut point of 88 cm (currently recommended for women) the specificity drops markedly. (Too many false positives for insulin resistance.) “Abdominal obesity is overestimated in women.”

CONCLUSION

Waist circumference is a simple tool to exclude insulin resistance. If the patient has a circumference under 100 cm, he or she is very unlikely to have insulin resistance and hyper-insulinemia. Circumferences above 100 cm (40 inches) may, or may not, be related to insulin resistance.

BMJ June 11, 2005: 330: 1363-64 Original investigation, first author Hans Wahrenberg, Karolinska Institute, Stockholm, Sweden.

6-4 SENSITIVITY, SPECIFICITY, AND PREDICTIVE VALUES: A Review By The Editor

SENSITIVITY

Start by considering only subjects who have the disease, in this case insulin resistance.

Sensitivity of the 100 cm test for men:

Sensitivity of the test = % of subjects who have the disease (insulin resistance) who have a positive test (waist 100 cm and above)—ie, the true positive %.

Insulin resistance present (n = 284)

(The test) (disease present)

Waist 100 cm and above (true positive test) 277

Waist under 100 cm (negative test) 7

Total 284

The sensitivity of the test for men = 277/284 = 98% (the true positive %)

By the same calculations the sensitivity of the test in women = 63%.

SPECIFICITY

Start by considering only subjects who do not have the disease—are not insulin resistant (are insulin sensitive)

Specificity of the 100 cm test for men:

Specificity of the test = % of subjects who do not have the disease (are insulin sensitive) who have a negative test (waist under 100 cm—the true negative %).

Insulin sensitive (n = 469)

(The test) (disease absent)

Waist 100 cm and above (false positive test) 176

Waist under 100 cm (true negative test) 293

Total 469

Specificity of the test for men =true negative % = 293/469 = 63%

By the same calculations the specificity of the test in women = 97%

PREDICTIVE VALUE OF POSITIVE TESTS (POSITIVE PREDICTIVE VALUE)

Start by conside