 |
||
| Departments | Articles | |
||
Tips from Other Journals
Use of Tamoxifen to Prevent Breast Cancer: A British Trial
The use of tamoxifen to prevent breast cancer is controversial. Many ethical issues and other concerns about exposing healthy women to an increased risk of endometrial cancer and vascular events have been overshadowed by the impressive reductions in breast cancer reported by an American trial, the National Surgical Adjuvant Breast and Bowel Project (NSABP-P1). In that four-year clinical trial, tamoxifen appeared to prevent approximately one half of invasive and noninvasive cancers in women at increased risk of breast cancer. Powles and associates report the results of an interim analysis of a British study of tamoxifen for the prevention of breast cancer.
The study included women 30 to 70 years of age with an increased risk of breast cancer because of a family history. They were randomized to receive either 20 mg of tamoxifen daily or placebo. The preliminary results reported in this study are based on 2,471 women. Median follow-up was 70 months. A total of 156 patients completed the eight years of therapy; 877 patients stopped therapy because of side effects (320 women receiving tamoxifen and 176 women receiving placebo) or for other reasons.
The frequency of breast cancer was found to be the same in the tamoxifen and placebo groups. In the tamoxifen group, breast cancer developed in 34 women compared with 36 women in the placebo group. Nulliparous women were found to have a twofold increase in the risk of breast cancer compared with women with children. Women who were receiving hormone replacement therapy at the time of entry into the study had an increased risk of breast cancer, but those who started hormone replacement therapy during the study had a decreased risk. However, after adjusting for confounding variables, there appeared to be no interaction between the use of hormone replacement therapy and any effect of tamoxifen in preventing breast cancer.
Four cases of endometrial cancer occurred in the tamoxifen group, compared with one case in the placebo group. Nine deaths occurred in the group treated with tamoxifen (four of which were from breast cancer), compared with six deaths in the placebo group (one of which was from breast cancer). Deep venous thrombosis occurred in four patients receiving tamoxifen and in two patients receiving placebo. Three patients receiving tamoxifen had pulmonary embolism compared with two patients receiving placebo.
The authors state that they were surprised to find no overall reduction in the occurrence of breast cancer in patients receiving tamoxifen. They note that one reason for the difference in the findings from the NSABP trial and their study may relate to study populations. Entry criteria for the British study were based mainly on a strong family history, with an associated risk of inheriting a gene that predisposes the women to breast cancer. Entry criteria for the NSABP study were primarily based on nongenetic risk factors. Another reason for the discordant findings may relate to the duration of follow-up. The median follow-up for the British study was nearly six years, compared with three to five years in the NSABP trial.
The authors conclude that there remains doubt about what the findings in the NSABP trial mean in terms of breast cancer prevention. Further study is necessary to characterize the subgroups in which the benefits of tamoxifen outweigh the risks and to justify the use of tamoxifen for the prevention of breast cancer.
ANNE D. WALLING, M.D.
Powles T, et al. Interim analysis of the incidence of breast cancer in the Royal Marsden Hospital tamoxifen randomised chemoprevention trial. Lancet July 11, 1998; 352:98-101.
Changes in Melanocytic Nevi from Ultraviolet Radiation
Exposure to UV radiation is linked to the development of nevi, and studies suggest that sunburns during childhood may be an independent risk factor for malignant melanoma. Recognition of changes in melanocytic nevi is important, and dermoscopy (magnification of the skin) has become a routine clinical method of examining these lesions. Because UV exposure can cause clinical, histologic and ultrastructural changes in melanocytic nevi, Hofmann-Wellenhof and associates investigated whether exposure to two minimal erythema doses of UV radiation induces changes in the dermoscopic evaluation of melanocytic nevi.
Four volunteers with a total of 15 melanocytic nevi participated in the study. Each nevus and 2 cm of adjacent skin were exposed to two minimal erythema doses of UV radiation. The lesions were then examined for differences in their dermoscopic appearance after exposure. The nevi were graded before the UV exposure and again three, seven, 14 and 28 days after exposure using a visual analog scale.
Three days after UV irradiation, the borders of the nevi were more faded, the nevi were more darkly pigmented and the hypopigmented areas were significantly less prominent. The pigment network structures were less prominent and more faded. No significant changes were noted in asymmetry, erythema in the nevus, telangiectasia, regularity of pigment network or presence of brown-black globules.
The changes in the magnified images were most prominent on the seventh day. The borders of the nevi were more faded, more dark pigmentation was present and hypopigmented areas were significantly less prominent. The number and intensity of the brown-black globules were increased.
On the 14th day after UV irradiation, the nevi were still darker brown, and hypopigmented areas were fading. On the 28th day, only the hypopigmented areas were still less prominent than before UV irradiation.
It is likely that the increase in pigmentation and in brown-black globules reflect induction of melanin synthesis. The changes observed in the nevi are in accordance with histopathologic findings of nevi excised in summer months, when sun exposure is maximal.
Dermoscopic evidence of an increase in brown-black globules and darkening of pigmentation in melanocytic nevi has been cited by some authors as a sign of malignant melanoma. Even with the use of 3 10 magnification, the black-brown globules are sometimes indistinguishable from the black dots that are one of the most specific dermoscopic criteria for melanoma. In contrast, a faded border, discrete pigment network, low amount of hypopigmented area and regularity of pigment network are considered to be dermoscopic criteria for benign nevi.
The authors conclude that low-dose UV radiation, such as that occurring with a sunburn, can induce transient changes in melanocytic nevi. These changes are primarily due to increased pigmentation. Although many of the changes observed in the nevi may be clinically interpreted as benign, the hyperpigmentary changes may be misidentified as malignant melanoma. Because of this possibility, the authors recommend that melanocytic nevi in patients with a recent history of sunburn be reexamined one month later to assure accurate diagnosis and avoid unnecessary biopsy.
BARBARA APGAR, M.D., M.S.
Hofmann-Wellenhof R, et al. Ultraviolet radiation of melanocytic nevi: a dermoscopic study. Arch Dermatol July 1998;134:845-50.
EDITOR'S NOTE: Physicians are increasingly learning that the colposcope can be a tool for evaluating lesions in areas other than the lower genital tract. The availability of colposcopes in family physicians' offices allows more accurate assessment of skin lesions. Magnification of the lesion on a video screen affords the opportunity for teaching and discussion. In addition, the images can be captured by videotape or photograph to facilitate monitoring changes in a lesion's appearance.
B.A.
Timing of Administration of Tissue Plasminogen Activator
Patients with acute myocardial infarction who are treated with tissue plasminogen activator (t-PA) or streptokinase have decreased rates of morbidity and mortality when therapy is given between five and 12 hours after acute symptom presentation. Less benefit or even questionable benefit is seen in patients treated with either thrombolytic regimen more than 12 hours after the onset of acute coronary symptoms. Goldberg and associates used an observational database derived from the second National Registry of Myocardial Infarction (NRMI) to investigate the use and timing of thrombolytic therapy after acute myocardial infarction.
The NRMI database was used because published randomized trials have become increasingly generalized to patients treated with thrombolytic therapy in the community rather than a selected controlled trial population. The registry is a voluntary database of patients hospitalized with documented acute myocardial infarction since 1990; data from NRMI-2, which studied patients from June 1994 to April 1996, were used for this study. Twenty-six percent of all acute medical-surgical hospitals in the United States participated in the database.
Of the 71,253 patients with acute myocardial infarction who were treated with t-PA, 7 percent received treatment within one hour of acute pain onset; 32 percent were treated within one to two hours; and 36 percent were treated within two to four hours. Treatment was administered within four to six hours of acute pain onset in 12 percent of patients and within six to 12 hours in 9 percent of patients; nearly 4 percent of patients received treatment sometime thereafter. Those patients who presented to the hospital sooner were more likely to receive t-PA treatment earlier. A clear association was noted between earlier treatment with t-PA and a lower risk of dying during the acute hospitalization.
The authors conclude that early administration of thrombolytic therapy after acute myocardial infarction improves outcome, and that patient delay in seeking medical care is a major impediment to the receipt of thrombolytic therapy. Patients should be educated not to delay seeking treatment so that early use of thrombolytic therapy can be ensured in appropriate patients.
RICHARD SADOVSKY, M.D.
Goldberg, et al. Impact of time to treatment with tissue plasminogen activator on morbidity and mortality following acute myocardial infarction (the second National Registry of Myocardial Infarction). Am J Cardiol August 1, 1998;82:259-64.
Evaluation of the Permanence of Hair Removal by the Ruby Laser
Several laser products have been commercialized for hair removal, but their efficacy has been questioned. Dierickx and associates evaluated the permanence of hair removal by normal-mode ruby laser treatment. This study was a two-year follow-up assessment of subjects in a previous study of a single laser treatment for hair removal on the thighs and backs of 13 volunteers.
The current study included seven of the original 13 participants. Baseline hair counts were obtained from eight test sites on each subject, and before laser exposure, one half of the test sites were shaved and one half were epilated with cold wax. Sites were irradiated with a normal-mode ruby laser at fluences of 0 (unexposed control), 30, 40 and 60 J (joules) per cm2. Terminal coarse hair counts were performed one, three, six, 12 and 24 months after laser exposure.
Follow-up revealed that four of the seven subjects still exhibited hair loss in the laser-treated sites two years after treatment; three had complete or nearly complete hair regrowth. All seven subjects showed no significant change in terminal hair counts six months, one year and two years after laser exposure, suggesting that six months' follow-up may be sufficient for assessing the outcome after laser therapy for hair removal. Neither pigment changes nor scarring was noted in any subject at the one- and two-year follow-up visits.
Sites treated with the highest laser fluence had the greatest hair loss. Compared with shaved and epilated control sites, laser-treated sites demonstrated significant hair loss at six months for all fluences at both shaved and epilated sites. At one- and two-year follow-up, significantly less hair appeared only in the shaved sites for all fluences compared with the untreated control site.
Terminal and vellus-like (miniaturized) hairs were examined histologically. The total number of hairs was identical in the control and laser-treated sites. However, in the laser-treated sites, a reduction in large terminal hairs and a reciprocal increase in small vellus-like hairs had occurred. The average hair shaft diameter measured on histologic sections also decreased after laser treatment. There were no signs of fibrous tracts, and normal-appearing sebaceous glands were still present around the villus hairs.
The findings indicate that permanent loss of terminal hair can result from a single exposure to high-fluence ruby laser therapy. The pulses generated by this particular laser were long enough to cause thermal coagulation and vaporization injury of hair follicles, leading to a growth delay in all subjects and permanent hair loss in some. This study suggests that the most important factor producing laser-induced alopecia is miniaturization of coarse terminal hair follicles to vellus-like hair follicles, resulting in finer and softer hair.
The authors hypothesize that two responses occur after laser treatment: a delay in hair growth and permanent hair loss. They postulate that the sensitivity of hair follicles to laser pulses may vary with the hair growth cycle. Hairs resistant to permanent inactivation may be in the telogen stage at the time of treatment.
BARBARA APGAR, M.D., M.S.
Dierickx CC, et al. Permanent hair removal by normal-mode ruby laser. Arch Dermatol July 1998;134:837-42, and Tope WD. A hair's breadth closer? [Editorial]. Arch Dermatol July 1998;134:867-9.
EDITOR'S NOTE: In an accompanying editorial, Tope notes the difficulty of obtaining accurate data on the efficacy of hair removal methods. Many studies have been based on clinical observations or patient satisfaction, without good follow-up data. Although the number of patients in this study was small, this study represents an attempt to define the safety and efficacy of hair removal by the pulse laser. This method may offer women with androgen-excess syndromes hope for permanent hair removal, which has been impossible to achieve with other therapies.
B.A.
Use of Montelukast for Exercise-Induced Bronchoconstriction
The etiology of exercise-induced bronchoconstriction in patients with asthma is not well understood but is thought to relate to cooling and drying of the airways, which stimulates the release of inflammatory mediators such as leukotrienes C4, D4 and E4. Leff and colleagues evaluated the effects of montelukast, a leukotriene receptor antagonist, on controlling airway hyperreactivity in response to exercise and methacholine challenge in patients with mild asthma.
The 12-week, double-blind, placebo-controlled study included 110 patients from 15 to 45 years of age (average age: 25 years) with at least a one-year history of asthma. Patients were nonsmokers and were receiving only inhaled beta agonists for therapy. All patients demonstrated a 20 percent or more decrease in forced expiratory volume in one second (FEV1) after a methacholine challenge and after a standardized exercise challenge on two occasions. None of the patients had used corticosteroids, long-acting antihistamines, theophylline, oral or long-acting beta adrenergic agonists or inhaled anticholinergics in the month before the study.
During the study, 54 patients received 10 mg of montelukast daily, and 56 received placebo. Patients were given an exercise challenge at baseline and four, eight and 12 weeks after beginning treatment; a methacholine challenge was given at baseline and at weeks 4 and 12. After 12 weeks of treatment, all patients received placebo for two weeks, and the exercise and methacholine challenges were again performed. The exercise challenge consisted of six minutes on a treadmill while inhaling compressed dry air at room temperature. Spirometry was performed at frequent intervals after the exercise challenge, beginning at zero minutes and then at five, 10, 15, 30, 45 and 60 minutes after exercise. If the FEV1 had not returned to within 5 percent of the pre-exercise value by 60 minutes, measurements were also obtained at 75 minutes and, if necessary, at 90 minutes after exercise. This process quantified the extent and duration of postexercise bronchoconstriction. For the methacholine challenges, the concentration of methacholine required to decrease the FEV1 by 20 percent was measured. After the 12 weeks of treatment, the patients were asked to evaluated the control of their asthma using a seven-point scale.
The mean FEV1 before exercise at baseline and at week 12 was similar in both groups. In response to the exercise challenges, patients receiving montelukast were found to have significantly greater protection from bronchospasm than patients receiving placebo. The treatment group also had significant improvement in the maximal decrease in FEV1 after exercise and in the interval between the maximal decrease in FEV1 and the return to within 5 percent of the pre-exercise FEV1. Two weeks after cessation of montelukast therapy, the mean values for FEV1 in the montelukast group were similar to those of the placebo group.
With the methacholine challenge, patients receiving montelukast required proportionately more double doses of methacholine than did the placebo group, although the difference did not reach statistical significance.
Scores on the self-assessment of symptoms after 12 weeks were much better in the montelukast group, with 73.1 percent characterizing the control of asthma as better. In contrast, 44.4 percent of patients in the placebo group described symptom control as better. The two groups did not differ significantly in the frequency of laboratory or clinical adverse events.
The authors conclude that montelukast at a daily dosage of 10 mg provides significant, consistent protection against exercise-induced bronchospasm. Unlike the short-acting inhaled beta agonist albuterol, montelukast does not appear to lead to the development of tolerance with prolonged use. Although the study demonstrated that montelukast offered no residual protective effect after cessation of therapy, no rebound worsening of symptoms was observed two weeks after discontinuation.
JEFFREY T. KIRCHNER, D.O.
Leff JA, et al. Montelukast, a leukotriene-receptor antagonist, for the treatment of mild asthma and exercise-induced bronchoconstriction. N Engl J Med July 16, 1998; 339:147-52.
New Treatment for Refractory Postpartum Hemorrhage
Postpartum hemorrhage is the major cause of maternal death and morbidity worldwide, accounting for at least 100,000 deaths and 20 million morbidities annually. Postpartum hemorrhage is commonly attributed to uterine atony following delivery. Initial treatment includes use of either intravenous or intramuscular oxytocin and ergometrine, plus supportive measures. Parenteral prostaglandin is administered if bleeding persists. Surgical options, such as ligation of the iliac artery or, as last resort, hysterectomy, are indicated if all other measures fail. Misoprostol, a prostaglandin E1 analog commonly used for peptic ulcer disease, has been adapted for use in obstetrics and gynecology. O'Brien and colleagues investigated the effectiveness of rectally administered misoprostol for postpartum hemorrhage unresponsive to conventional first-line management.
Fourteen women who required emergency management of severe postpartum hemorrhage were included in the study. Median age of the women was 31.5 years, median gestation was 39 weeks and parity ranged from zero to five. Initial management of all women was similar. Hospital protocol for treatment of postpartum hemorrhage was followed, consisting of resuscitation, bimanual uterine compression and administration of parenteral ergometrine and oxytocin unless contraindicated because of preeclampsia. However, if bleeding persisted despite these measures, 1,000 mg of misoprostol was administered rectally.
Bleeding was controlled in all 14 women, and sustained uterine contractions were achieved within three minutes of administration of misoprostol. Estimated blood loss ranged from 500 to 2,000 mL, and 11 of the 14 women required blood transfusion (2 to 4 units of blood). The mode of delivery varied. Of the 14 women studied, eight had spontaneous vaginal vertex deliveries, two had emergency cesarean deliveries, two had ventouse extractions, one an elective cesarean delivery and one a vaginal breech delivery. Ten mothers experienced complications of pregnancy, including preeclampsia, abruption, retained placenta and breech presentation.
The authors conclude that misoprostol provided rapid, effective treatment of postpartum hemorrhage. Misoprostol is inexpensive and does not elevate blood pressure, which is particularly significant in patients with preeclampsia. Parenteral prostaglandins have several disadvantages: they are more expensive than misoprostol, they must be protected from light and heat, and they must be injected into muscle or the myometrium. In addition, they can elevate blood pressure, cause pulmonary edema and exacerbate asthma. The authors anticipate that rectally administered misoprostol could become the treatment of choice in postpartum hemorrhage that is unresponsive to oxytocin and ergometrine.
ANNE D. WALLING, M.D.
O'Brien P, et al. Rectally administered misoprostol for the treatment of postpartum hemorrhage unresponsive to oxytocin and ergometrine: a descriptive study. Obstet Gynecol August 1998;92:212-4.
EDITOR'S NOTE: Few physicians who have treated a patient with intractable postpartum hemorrhage remain unscathed. For ethical and practical reasons, a formal, randomized, controlled clinical trial evaluating the use of rectal misoprostol in postpartum hemorrhage may not be possible, or it may take years to report conclusive results. In the meantime, this impressive study suggests a new adjunctive therapy that may be particularly useful in developing countries where resources are limited.
A.D.W.
Does Nonoxynol 9 Help Prevent STDs in Women?
Nonoxynol 9 has been used as a chemical spermicide since the 1950s. This chemical compound, through its disruption of epithelial cells, bacteria and viruses, also acts in vitro against a variety of pathogens, including Chlamydia trachomatis, Neisseria gonorrhoeae and Treponema pallidum. Evidence from in vitro studies suggests that nonoxynol 9 is able to inactivate the human immunodeficiency virus (HIV); however, clinical results have been conflicting. Roddy and colleagues performed a randomized, controlled trial to determine if nonoxynol 9 used before intercourse could reduce the rate of sexually transmitted diseases (STDs), including HIV, among women.
Study participants were HIV-seronegative female sex workers in Cameroon who were between 18 and 45 years of age and averaged at least four sexual partners per month. They were given a vaginal film that contained 70 mg of nonoxynol 9 or a placebo film that was identical in appearance. The 5 3 5-cm film was inserted vaginally before intercourse and dissolved in two to five minutes. In addition, all women were given plain latex condoms for their partners to use. Both the film and the condom were recommended for use with each sexual encounter.
The women were instructed to keep a log for recording the frequency of vaginal, oral and anal intercourse. They were also to include use of antibiotics, receipt of injections and use of other vaginal agents such as douches. Each woman also underwent monthly HIV testing, examination for genital and cervical ulcerations, and DNA-probe tests for gonorrhea and Chlamydia, which were performed by the researchers. The patients were also counseled again at each visit regarding the correct and continuous use of the vaginal film and the condoms.
A total of 1,170 women were randomized into the study and ultimately 941 completed the 12 months of follow-up. The average age of the participants was 26 years, and the prevalence of STDs was similar at baseline in both the control and the treatment groups. Compliance with the use of condoms and vaginal film ranged from 63 to 86 percent. During the study, there were 94 HIV seroconversions, including 48 in the treatment group and 46 in the placebo group. There were 111 cases of gonorrhea and 81 cases of Chlamydia infection in the placebo group compared with 114 cases of gonorrhea and 79 cases of Chlamydia infection in the nonoxynol 9 group. The rate of genital lesions in the treatment group was 42.2 percent compared with 33.5 percent in the placebo group. However, the presence of genital ulcerations did not predict HIV seroconversion, as has been reported in other studies.
The authors conclude that nonoxynol 9 does not provide additional protection from gonorrhea, Chlamydia or HIV infection beyond that provided by condoms. Furthermore, they state that there is a compelling need to develop better barrier methods that can be controlled by women for protection against sexually transmitted infections.
JEFFREY T. KIRCHNER, D.O.
Roddy RE, et al. A controlled trial of nonoxynol 9 film to reduce male-to-female transmission of sexually transmitted diseases. N Engl J Med August 20, 1998;339:504-10.
Transurethral Injections for Intrinsic Sphincter Deficiency
Up to one fourth of women between 15 and 60 years of age report symptoms of urinary incontinence. In most, incontinence results from a lack of support of the vesicourethral junction. Another major cause of stress incontinence is intrinsic sphincter deficiency caused by trauma, surgical injury, sympathetic nerve dysfunction or myelodysplasia. Conventional surgical treatments of intrinsic sphincter deficiency may result in detrusor problems leading to urinary retention, manifested as frequency, urgency and urge incontinence. Koelbl and colleagues assessed the short-term efficacy of transurethral injection of silicone microimplants in women with intrinsic sphincter deficiency.
Thirty-two women (median age: 64.3 years) who had intrinsic sphincter deficiency as the cause of stress incontinence were enrolled in the study. Twenty-eight of these women had undergone previous surgical procedures: 18 women had had one surgical intervention, and 10 had two previous surgeries for incontinence. Each woman underwent a comprehensive urogynecologic assessment before transurethral injection of silicone microimplants. The implants were placed under direct vision at 2-, 6- and 10-o'clock positions, 2 cm distal to the bladder neck. Each woman underwent full clinical and urogynecologic review after six and 12 months.
Immediately following surgery, all 32 patients reported continence of urine; urodynamic measurements validated the subjective reports. After six months, 24 patients (75 percent) still reported continence, validated by measurements. By 12 months, 19 women (59 percent) reported continence. For the entire group, the injections significantly increased maximal urethral closing pressures. Two patients developed urinary tract infections following surgery, and two patients required sling procedures three months after the injections failed. No other complications were noted during the study.
The authors conclude that this relatively simple new technique can provide complete resolution of incontinence. After 12 months, approximately 60 percent of patients in this study were subjectively and objectively continent. Although the benefit of the procedure may decrease over time, the authors believe it will provide a good treatment option, particularly in the elderly or in those patients who are unfit for major surgery.
ANNE D. WALLING, M.D.
Koelbl H, et al. Transurethral injection of silicone microimplants for intrinsic urethral sphincter deficiency. Obstet Gynecol September 1998;92:332-6.
The Minimally Abnormal Pap Smear: Predictor of Dysplasia?
The further evaluation of patients whose Papanicolaou smears show atypical squamous cells of undetermined significance (ASCUS) or low-grade squamous intraepithelial lesions (SIL) remains controversial. Selecting the optimal management strategy is complicated because the incidence of high-grade dysplasia in these women varies. Kobelin and colleagues conducted a retrospective chart review to determine the clinical predictors of significant dysplasia in patients with minimally abnormal Pap smears.
The charts of more than 400 women referred for colposcopy because of Pap smears interpreted as showing ASCUS or low-grade SIL were reviewed. Exclusion criteria included women who were pregnant or in the postpartum period and those with a history of in utero exposure to diethylstilbesterol, lower genital tract dysplasia or human immunodeficiency virus infection. All colposcopic examinations were conducted by the same examiner, and all specimens were examined by the same team of pathologists.
The median age of the study participants was 30 years. Sixty percent of the women were nulliparous, 17 percent were smokers and 32 percent used oral contraceptives. No cases of invasive cervical cancer were diagnosed at colposcopy. A total of 137 women (34 percent) had either low-grade SIL or high-grade SIL. Colposcopy confirmed low-grade SIL in 47 of the 203 women (23.2 percent) with SIL on Pap smear. An additional 42 women with low-grade SIL on Pap smear (20.7 percent) had high-grade SIL at colposcopy. Of the 203 women with ASCUS on Pap smear, low-grade SIL was diagnosed by colposcopy in 31 (15.3 percent), and high-grade SIL was diagnosed by colposcopy in 17 (8.4 percent).
The strongest independent predictor of histologic high-grade SIL at colposcopy was age younger than 35 years. These women were 2.6 times more likely than women older than 35 years to have histologic high-grade SIL by colposcopy. The next strongest predictor was low-grade SIL on Pap smear and a history of condyloma. Patients with low-grade SIL on Pap smear were 2.5 times more likely to have histologic SIL at colposcopy than patients whose Pap smears were reported as ASCUS. The subgroup of women with ASCUS who were 35 years or older had the lowest incidence of SIL at colposcopy (14 percent). The subgroup of women younger than age 35 with low-grade SIL on Pap smear had the highest incidence of SIL (51 percent) and high-grade SIL (24 percent) at colposcopy.
The authors conclude that colposcopy should be considered for use in all women with minimally abnormal Pap smears because of the high prevalence of SIL. However, the optimal approach to management of these women remains unknown. The authors suggest that since age and histologic status of the Pap smear are predictive of dysplasia, these factors could be used to identify patients at particularly high risk of cervical dysplasia.
ANNE D. WALLING, M.D.
Kobelin MH, et al. Incidence and predictors of cervical dysplasia in patients with minimally abnormal Papanicolaou smears. Obstet Gynecol September 1998;92: 356-9.
Triple Therapy: 10 vs. 14 Days for Eradication of H. pylori
Since consistently higher cure rates of Helicobacter pylori have been reported with the use of proton pump inhibitorbased triple therapy, this therapy is recommended over dual therapy. However, the ideal duration for triple therapy, from seven to 14 days, is still being evaluated. Studies showing more than than 90 percent H. pylori eradication using triple therapy for seven days have been reported. Fennerty and associates compared a 10-day treatment regimen with a 14-day treatment regimen in patients with H. pylori infection and active duodenal ulcer disease or a history of duodenal ulcer.
A total of 284 patients were randomized to receive 30 mg of lansoprazole, 250 mg of amoxicillin and 500 mg of clarithromycin twice daily for 10 or 14 days. The primary efficacy outcome was eradication of H. pylori, confirmed by histologic and culture results four to six weeks after completion of therapy. Because of various protocol violations, only 236 patients were included in the final analysis.
H. pylori was eradicated in 103 patients (84 percent) in the 10-day treatment group and in 96 patients (85 percent) in the 14-day treatment group. Even in the worst-case scenario, in which patients who failed to return for follow-up were considered to have persistent infection, the H. pylori eradication rates were 81 percent in the 10-day group and 82 percent in the 14-day treatment group. No significant differences were found between the two groups in overall treatment outcomes. Of the 228 patients with an active ulcer, 166 patients did not have an ulcer at the final endoscopy, conducted four weeks after the completion of therapy. No difference in ulcer status was noted between the two treatment groups, and the same percentage of patients showed resolution or improvement of acute and chronic inflammation.
No differences were observed between the treatment groups in either the frequency or the mean severity of daytime or nighttime abdominal pain during the treatment period. There was, however, a significant observed difference in the frequency and mean severity of daytime abdominal pain during the follow-up period. Adverse events did not differ between treatment groups. The most commonly reported events were diarrhea and taste disturbance. Nine patients discontinued taking the study medication prematurely, at least in part because of drug-related adverse events.
The results of this study are similar for the 10- and 14-day triple regimens. The treatment groups were found to be comparable based on cure rates of H. pylori infection, resolution or improvement of gastritis, frequency and mean severity of daytime or nighttime abdominal pain, and tolerability. The authors conclude that triple therapy twice daily for 10 days results in high rates of H. pylori eradication, tolerability and compliance.
BARBARA APGAR, M.D., M.S.
Fennerty MB, et al. A comparison of 10 and 14 days of lansoprazole triple therapy for eradication of Helicobacter pylori. Arch Intern Med August 10, 1998;158: 1651-6.
Should Relatives Be Witnesses to Resuscitation Attempts?
Family members are usually excluded from resuscitation attempts because it is believed that witnessing invasive procedures and aggressive interventions would be distressing to them. In addition, the presence of family members is thought to impair the performance of clinical staff. Robinson and colleagues conducted a pilot study to see if witnessing a resuscitation attempt had an adverse effect on family members who remained with the patient during the attempt.
During a 15-month period, relatives of patients undergoing resuscitation following cardiac arrest or multiple trauma at a British hospital were included in the study. A sealed envelope was used to randomize relatives who would be eligible to remain with the patient during resuscitation attempts. However, those randomized to the control group were not given this option. Both groups of relatives were accompanied at all times by an experienced staff member who explained the resuscitation procedures according to standardized protocols.
Bereaved relatives were interviewed at one and six months after the resuscitation attempt, and their psychological status was established using five questionnaires. The results of the questionnaires were available, on average, three and nine months after the event.
Of the 25 patients resuscitated, 13 were in the witnessed resuscitation group and 12 were in the control group. All of the patients in the control group died, but three of the patients in the witnessed resuscitation group survived. Thirty-four relatives were present in the witnessed resuscitation group, but only the person most closely related to the patient was included in the analysis. Full psychologic testing at six months was obtained for eight bereaved relatives who witnessed resuscitation attempts and for 10 relatives from the control group. At three and nine months, the median scores for the witnessed resuscitation group were lower than those of the control group, suggesting decreased levels of anxiety, depression, intrusive imagery, grief and post-traumatic avoidance behavior. All of the relatives who remained with the patient during the resuscitation attempt expressed satisfaction with the decision.
The trial was terminated early because the randomization process was at risk of being compromised when the clinical team became convinced that allowing relatives to remain during resuscitation was beneficial. The authors emphasize that relatives should be offered the choice to remain with the patient and that if they choose to remain, they must be supported and offered explanations throughout the procedure.
The authors conclude that routine exclusion of relatives from resuscitation is no longer appropriate.
ANNE D. WALLING, M.D.
Robinson SM, et al. Psychological effect of witnessed resuscitation on bereaved relatives. Lancet August 22, 1998;352:614-7.
Zinc Can Reduce Pediatric Respiratory Infections
Acute lower respiratory infections, particularly pneumonia, are responsible for approximately one third of all childhood deaths in developing countries. Malnutrition increases both the severity and the incidence of these infections. Malnourished children have a decreased immunologic capacity, a problem attributable to zinc deficiency. Sazawal and colleagues conducted a double-blind, randomized trial to evaluate the effectiveness of zinc supplementation in improving immune status and reducing the incidence of lower respiratory infections in children.
Children enrolled in the study were between six and 35 months of age, were of low socioeconomic background and presented to a clinic in India for treatment of acute diarrhea. The children were assigned to receive either a 5-mL multivitamin preparation daily or 5 mL of zinc daily for six months. During episodes of acute diarrhea, children in the zinc group received 10 mL per day. Anthropometric assessments were conducted at baseline and at monthly intervals. Plasma zinc levels were measured at baseline and also after 120 days of supplementation. Zinc deficiency was defined as a plasma level less than 60 µg per dL (9.2 µmol per L). Information about the presence of fever, cough, difficulty in breathing and feeding was recorded every fifth day. Acute lower respiratory infection was defined as cough plus one or more of the following: tachypnea, temperature of 101°F (38.3°C) or lower chest wall retractions.
Of the 609 children enrolled in the study, 311 were assigned to the control group, and 298 were assigned to the zinc group. Plasma zinc levels were similar between the two groups at baseline. However, after 120 days of supplementation, zinc levels were significantly higher in the group receiving the zinc supplement. The number of episodes of acute lower respiratory infection was lower in the zinc group (0.19 episodes per child per year) when compared with the control group (0.35 episodes per child per year). In a corrected analysis using the World Health Organization's criteria for respiratory disease, zinc supplementation resulted in an 11 percent decrease in these infections. The reductions tended to be greater in children older than 11 months and in boys, but the differences were not statistically significant.
The authors conclude that zinc supplementation is effective in reducing the morbidity associated with acute lower respiratory infections in infants and children, possibly by improving immune status. Further research is needed to determine whether dietary supplements of zinc and other micronutrients, along with additional dietary modifications or fortification of staple foods, are a means of improving health in children.
GRACE BROOKE HUFFMAN, M.D.
Sazawal S, et al. Zinc supplementation reduces the incidence of acute lower respiratory infections in infants and preschool children: a double-blind, controlled trial. Pediatrics July 1998;102:1-5.
Treatment and Prognosis in Guillain-Barré Syndrome
Guillain-Barré syndrome (GBS) is the leading cause of flaccid paralysis in western countries. It is a self-limiting, autoimmune, neurologic disease that presents as rapidly progressive symmetric limb weakness, loss of tendon reflexes and autonomic dysfunction. While the cause of GBS is still unclear, various infections have been established as triggers. The most frequent antecedent pathogen is Campylobacter jejuni, but Cytomegalovirus and various vaccinations have also been implicated in the etiology of GBS. A review by Hahn discusses the epidemiology, progression and treatment of the disease.
Patients require supportive therapy that is appropriate to the extent of their symptoms. Approximately one third of patients with GBS require intensive care and ventilatory support. Autonomic dysfunction and hemodynamic instability are common, requiring vigilance for the potential complications of thromboembolism, cardiac dysfunction, ventilatory failure and oropharyngeal weakness. Patients may need total supportive care, including pain and symptom control, skin and bowel care, hydration and nutritional support, and psychologic services. Intravenous steroid therapy has not been shown to be beneficial in the treatment of GBS.
Plasma exchange can be effective if used within two weeks of symptom onset. At least three controlled trials have reported a more rapid return to ambulation, fewer days on ventilation support and in the hospital, and lower costs for patients treated with plasma exchange. After an initial improvement, approximately 10 percent of patients experience secondary relapse within two weeks of starting plasma exchange. Current recommendations range from two plasma exchange treatments for mild GBS to five treatments for more severe cases. Plasma exchange is expensive and has significant risks, particularly in hemodynamically unstable patients.
Intravenous IgG therapy, usually given in a dosage of 0.4 g per kg for five days, has a lower risk and is easier to administer than plasma exchange. Two studies have concluded that IgG therapy yields results comparable to those of plasma exchange and that combining the two therapies does not confer a significant advantage. Relapses also occur in approximately 10 percent of patients treated with IgG therapy.
The authors conclude that although the management of GBS can be challenging and may require intensive therapy, the prognosis is usually favorable. The progression of the disease is highly variable. Approximately 75 percent of cases reach maximal dysfunction within two weeks, and 94 percent reach it within four weeks. Symptoms resolve gradually. In one series, 70 percent of patients reported full recovery within one year; 22 percent of patients were unable to run, and 8 percent could not walk without assistance. The percentage of patients who die from GBS varies among studies but is approximately 7 percent. Patients who are older than 60 years are at greatest risk of death and serious long-term complications. A preceding diarrheal illness and rapid disease progression are also associated with poor prognosis.
ANNE D. WALLING, M.D.
Hahn AF. Guillain-Barré syndrome. Lancet August 22, 1998;352:635-41.
Screening for Rotavirus in Children with Diarrhea
Diarrhea accounts for 9 to 13 percent of hospitalizations in children younger than five years of age. Epidemiologic characteristics of rotavirus, such as winter seasonality, predisposition to rotavirus infection in children four to 23 months of age and geographic migration of rotavirus epidemics, have been used to estimate that this infection causes one third of all pediatric hospitalizations associated with diarrhea. Parashar and associates conducted a retrospective analysis to assess the morbidity from diarrhea and to estimate the disease burden of rotavirus to determine the potential benefit of a rotavirus vaccine.
Data on approximately 1 million children up to six years of age who were enrolled in four health maintenance organizations were transmitted to the Centers for Disease Control and Prevention. Retrospective analysis of data from October 1992 through September 1994 included all events for which a diagnostic code specific to diarrhea was listed. Diarrhea was associated with 6.3 percent of all hospitalizations among children at the four sites. The majority of hospitalizations related to diarrhea (65.7 percent) were of unknown etiology. Rotavirus was specifically recorded as the cause in only 2.7 percent of hospitalizations. Fifty-seven percent of all diarrhea-associated hospitalizations were reported among children from four to 23 months of age, the most common age for diarrheal infections caused by rotavirus. Each year of the study period, the number of hospitalizations and visits to the emergency department associated with diarrhea peaked in the winter months.
The authors' findings indicate that diarrhea was an important cause of morbidity among these children. Further studies are needed to determine the exact burden of rotavirus on diarrheal illness, but the seasonality and age distribution of diarrheal illnesses hint strongly of the morbidity of rotavirus in this population. Laboratory-confirmed data on rotavirus infections are also needed to determine the potential cost-effectiveness and impact of rotavirus vaccines.
RICHARD SADOVSKY, M.D.
Parashar UD, et al. Epidemiology of diarrheal disease among children enrolled in four west coast health maintenance organizations. Pediatr Infect Dis J July 1998: 17;605-11.
EDITOR'S NOTE: A rotavirus vaccine (Rota Shield) has recently been recommended for approval by a panel from the U.S. Food and Drug Administration. The vaccine protects against the four most prevalent strains of rotavirus that cause symptoms. Administration of three doses is recommended at two, four and six months of age. The first dose should be given before six months of age because of the higher incidence of fever when the vaccine is first administered to children older than this age. Immunogenicity of the vaccine is not affected by the simultaneous administration of diphtheria, tetanus, pertussis, Haemophilus influenzae type B (DPT-Hib) vaccines, oral poliovirus vaccine (OPV), inactivated poliovirus vaccine (IPV) or hepatitis B (HepB) vaccine. The American College of Infectious Disease Pediatricians supports the use of this vaccine. The American Academy of Family Physicians is reviewing the trial data before making a final determination about including this vaccine among those recommended for use in children.
R.S.
Carcinoembryonic Antigen as a Tumor Marker for Gastric Cancer
Carcinoembryonic antigen (CEA) is widely used for establishing diagnosis, staging, detection of recurrence and prognosis in patients with colorectal cancer. The value of CEA as a tumor marker for gastric carcinoma is less clear. Tachibana and associates measured preoperative CEA levels in patients with resectable gastric cancer and evaluated the association between serum CEA levels and clinicopathologic features.
A total of 196 consecutive patients who underwent surgery for gastric carcinoma were included in the study. Of these patients, 138 had gastrectomy (70.4 percent), 54 had total gastrectomy (27.6 percent) and four had wedge resection (2.0 percent).
Serum CEA levels were above normal (range: 5.2 to 570 ng per mL [5.2 to 570 µg per L]) in 29 (14.8 percent) of the 196 patients. Of the 100 patients with early gastric cancer confined to the submucosal layer, only seven (7.0 percent) had positive CEA levels (more than 5.0 ng per mL [5.0 µg per L]). Of the 96 patients with advanced cancer, 22 (22.9 percent) had high CEA levels.
The cumulative three- and five-year survival rates in patients who had elevated CEA levels were significantly lower than those for CEAnegative patients. Three and five-year cumulative survival rates in CEApositive patients were 39.6 percent and 31.7 percent, respectively. In contrast, three- and five-year survival rates in CEAnegative patients were 83.0 percent and 77.3 percent, respectively.
The authors conclude that preoperative serum CEA levels provide predictive information in determining tumor stage and prognosis in patients with potentially resectable gastric cancer.
RICHARD SADOVSKY, M.D.
Tachibana M, et al. Serum carcinoembryonic antigen as a prognostic factor in resectable gastric cancer. J Am Coll Surg July 1998;187:64-8.
Montelukast for Management of Persistent Asthma
Montelukast sodium is a leukotriene receptor antagonist recently labeled by the U.S. Food and Drug Association for oral prophylaxis and ongoing treatment of asthma in adults and in children who are at least six years old. The previously released leukotrienes, zileuton and zafirlukast, are not labeled for use in children younger than 12 years. Drugs in the leukotriene modulator class decrease eosinophil migration, mucus production and airway wall edema. Consultants for the Medical Letter on Drugs and Therapeutics reviewed the current data on the leukotriene antagonists.
One double-blind, 12-week trial in adults with intermittent or persistent asthma indicated that 10 mg of montelukast taken at bedtime improved the morning forced expiratory volume in one second (FEV1) better than placebo. Use of montelukast also decreased nocturnal awakenings from asthma and decreased use of a short-acting beta agonist inhaler by 25 percent compared with a 10 percent decrease with placebo. The beneficial effects were near maximum within 24 hours after the first dose and were maintained for the course of the trial. An eight-week trial demonstrated similar beneficial results. Montelukast is not recommended for use alone as prophylaxis or treatment of exercise-induced bronchospasm.
In a 12-week trial, 10 mg of montelukast taken daily was less effective than inhaled beclomethasone, 200 µg twice daily, in increasing FEV1 in patients with moderate asthma. In a study of patients whose asthma was not adequately controlled with inhaled beclomethasone, the addition of montelukast proved more effective than the addition of placebo in improving lung function and decreasing symptoms. Adults with asthma that was well controlled with inhaled corticosteroids were able to decrease their corticosteroid dosage by 47 percent, compared with a 30 percent reduction with placebo. Furthermore, the combination of inhaled or oral corticosteroids and montelukast was shown to increase the FEV1 by 8.5 percent compared with a 1.7 percent decrease with placebo.
Montelukast is rapidly absorbed from the gastrointestinal tract, reaches a peak in three to four hours and is excreted mainly in bile. Hepatotoxicity (reported with the use of zileuton) and Churg-Strauss vasculitis (reported with the use of zafirlukast) have not been reported with the use of montelukast. Use of higher-than-recommended dosages of montelukast does not appear to increase the incidence of adverse effects. In short-term studies, the frequency of adverse events with montelukast is similar to that of placebo. Montelukast appears to have fewer drug interactions than zafirlukast because it does not inhibit CYP-450 enzymes, so theophylline, warfarin, prednisone, estrogens or progestins may be used concurrently.
Montelukast is available in 10-mg tablets and 5-mg chewable tablets, which both carry a wholesale cost of $67 for a 30-day supply, according to Drug Topics Red Book. The recommended dosage is 5 mg in children ages six to 14 years and 10 mg in adults, taken daily in the evening without regard to food.
Montelukast is modestly effective in controlling mild to moderate persistent asthma and is approved for use in children. It is less effective than inhaled corticosteroids, but the combination of the two drugs can allow the patient to decrease the corticosteroid dosage by almost 50 percent. It is more convenient and has fewer drug reactions than zafirlukast or zileuton.
BARBARA APGAR, M.D., M.S.
Medical Letter consultants. Montelukast for persistent asthma. Med Lett Drugs Ther July 17, 1998;40(1031):71-3.
Survival Rates in Men With Carcinoma of the Breast
Carcinoma of the breast is uncommon in men; the incidence is less than 1 percent of the incidence of breast cancer in women. For this reason, less clinical information is available regarding prognosis, treatment and outcomes for men. Interestingly, the survival rate of men with breast cancer is lower than that of women with breast cancer. Donegan and colleagues performed a retrospective case review of men in Wisconsin who were diagnosed with breast cancer to determine treatment, survival rates and prognostic factors.
Data were collected from the tumor registries of 18 hospitals in nine cities in eastern Wisconsin. The cancer programs at these hospitals were approved by the Commission on Cancer of the American College of Surgeons. The cases of male breast cancer included in this study were diagnosed between 1953 and 1995. A total of 221 cases were identified, and acceptable data were available on 215 patients. The men ranged in age from 32 to 90 years, with an average age of 65.4 years. The symptoms lasted an average of 10 months. Tumors were identified in the left breast in 127 men and in the right breast in 87 men. There was also one patient with bilateral disease. The type of cancer that was most commonly observed was invasive ductal carcinoma. The mean size of the breast mass was 2.6 cm and, although location was unspecified in most cases, direct involvement of the nipple was noted in 42 patients. Gynecomastia was occasionally observed, and several patients had a prior history of trauma to the chest.
In the 105 patients for whom survival data were available, the mean follow-up time was 45 months. Of the 110 patients who were known to have died, 64.5 percent died of breast cancer, 20 percent died of an unidentified cause, 8 percent died of other malignancies and 7 percent died of cardiovascular disease. Observed survival for all 215 patients was approximately 50 percent at five years and 24 percent at 10 years. Factors that affected survival included TNM staging at the time of diagnosis, including the involvement of four or more axillary lymph nodes. A positive estrogen and progesterone receptor status was associated with a better prognosis. The most common course of treatment was modified radical mastectomy, followed by chest wall irradiation. Radiation did not improve observed survival or survival when patients were stratified for lymph node status. Systemic adjuvant therapy plus hormonal therapy benefited only patients with positive axillary lymph nodes who were estrogen-receptor positive.
The authors conclude that men with breast cancer have clinical presentations, trends in treatment and prognostic factors similar to those in women with breast cancer. Although the study spanned 42 years, one half of the patients were diagnosed in the last 10 years. This is believed to reflect a proliferation of tumor registries with better reporting systems. The poorer survival rate in men has been attributed to older age and more advanced disease at the time of diagnosis, inappropriate staging, anatomic factors and high mortality from comorbid disease. Earlier diagnosis, surgical intervention and, when appropriate, systemic adjuvant therapy appear to improve long-term survival.
JEFFREY T. KIRCHNER, D.O.
Donegan WL, et al. Carcinoma of the breast in males. Cancer August 1, 1998;83:498-509.
Pyelonephritis in Pregnancy: Three Antibiotic Regimens
Acute pyelonephritis is a common complication of pregnancy and typically necessitates hospital admission. Therapy traditionally consists of inpatient administration of intravenous antibiotics, but recent studies have shown that intramuscular ceftriaxone can be an effective outpatient regimen in selected patients. Wing and colleagues conducted a randomized, two-center trial to compare the effectiveness of intramuscular ceftriaxone with two traditional intravenous antibiotic regimens in the treatment of acute pyelonephritis during pregnancy.
Pregnant women who presented to the emergency department with acute pyelonephritis were eligible for the study if they were at less than 24 weeks' gestation and had no contraindications to the study medication. Patients were randomly assigned to receive one of three drug regimens. The first group received 2 g of ampicillin intravenously every four hours and 1.75 mg per kg of gentamicin intravenously every eight hours following an initial dose of 2 mg per kg. The second group received 1 g of cefazolin intravenously every eight hours. The third group received two 1-g doses of ceftriaxone intramuscularly 24 hours apart, followed by 500 mg of oral cephalexin every six hours. Acetaminophen was administered, and cooling measures were initiated as needed to reduce fever and relieve discomfort. Patients were hospitalized until they were afebrile for 48 hours. On discharge, all patients were given a 10-day course of cephalexin, in a dosage of 500 mg four times daily, followed by 100 mg of nitrofurantoin daily for the remainder of the pregnancy and for six weeks postpartum.
Of the 179 patients enrolled in the study, 62 were given ampicillin-gentamicin, 58 were given cefazolin and 59 were given ceftriaxone. Baseline characteristics were similar in each group. The uropathogen most commonly found was Escherichia coli, present in 76.5 percent of the patients overall. Individual responses to therapy varied widely, but there were no significant differences across the three treatment groups. Patients taking ampicillin-gentamicin were afebrile, on average, after 10 hours, whereas patients in the other two groups were afebrile after 12 hours. The number of days to resolution of costovertebral angle pain and the length of hospital stay were also similar among the three groups. Four patients treated with cefazolin had prolonged fever compared with six patients in each of the other two groups.
Urine cultures were obtained from 149 of the patients two weeks after initial treatment. Four women in the ampicillin-gentamicin group, three women in the cefazolin group and one woman in the ceftriaxone group had positive results for organisms. Ten women had positive cultures later in their antepartum course, and 10 others developed recurrent pyelonephritis. Birth outcomes did not differ across groups. The average birth weight was 3,274 g (7 lb, 3 oz), and 11 (6.9 percent) births were premature.
The authors conclude that the three treatment regimens were equally safe and effective in treating pyelonephritis in pregnancy before 24 weeks' gestation. Significant savings could result from the selective use of ceftriaxone, as it can be administered on an outpatient basis.
ANNE D. WALLING, M.D.
Wing DA, et al. A randomized trial of three antibiotic regimens for the treatment of pyelonephritis in pregnancy. Obstet Gynecol August 1998;92:249-53.
Can Serologic Testing Confirm Cure of H. pylori Infection?
Serum IgG antibody tests for Helicobacter pylori remain positive months after treatment. Currently, paired serologies can be used to determine cure, or a patient can undergo urea breath tests or gastric biopsy. If symptoms recur, even a year later, the physician is left wondering if the patient has persistent or recurrent disease. Feldman and colleagues conducted a prospective study to determine the role of serologic testing in confirming cure of H. pylori infection more than one year after treatment.
Ten men and 13 women were included in this study; none had a history of peptic ulcer disease or chronic upper gastrointestinal tract symptoms. At baseline, H. pylori serum antibody tests were positive in each patient, as were gastric biopsies. Each patient was then treated with a two-week course of 524 mg of bismuth subsalicylate four times daily, 500 mg of tetracycline four times daily and 250 mg of metronidazole four times daily. Serum antibodies were retested, and mucosal biopsies of the gastric body were performed one, three and 18 months after completion of therapy. Biopsies of the gastric antrum were performed at three and 18 months. Cure was defined as absence of H. pylori organisms on both gastric biopsies at 18 months. Persistence was defined as the presence of organisms on biopsy specimens 18 months after therapy.
H. pylori infection was cured in 65 percent of the patients in the study approximately 18 months after antimicrobial therapy. In the remaining eight patients, infection persisted. Even in these patients, there was a 50 percent decrease in gastritis severity score at one month, but this was followed by an increase in severity over subsequent months. In the eight patients with persistent infection, H. pylori serum antibody levels did not change significantly over the course of the study. H. pylori seroconversion from positive to undetectable serum antibody levels at 18 months had a sensitivity of 60 percent for diagnosing cure of H. pylori infection and a specificity of 100 percent.
The authors conclude that a patient who was treated for H. pylori infection more than one year previously can reliably be considered cured if serum H. pylori antibody levels change from positive (pre-therapy) to negative. This serum test should probably be the test of first choice, since a negative test result would preclude the need for a urea breath test or gastric biopsy in most patients. If the patient has not seroconverted, the test result should be considered nonspecific, and a urea breath test or gastric biopsy should be done to confirm either cure or persistent infection. A serology test remains insensitive for confirmation of cure during the first three months after treatment.
GRACE BROOKE HUFFMAN, M.D.
Feldman M, et al. Role of seroconversion in confirming cure of Helicobacter pylori infection. JAMA July 22, 1998; 280:363-5.
More Effective Regimens for Emergency Contraception
The most commonly used emergency contraceptive is the Yuzpe regimen, which combines 100 mg of ethinyl estradiol with either 0.5 mg of levonorgestrel or 1.0 mg of norgestrel, in two doses over 12 hours. This regimen prevents approximately 75 percent of pregnancies if initiated within 72 hours of unprotected intercourse; however, it frequently causes nausea and vomiting. Levonorgestrel alone has also been marketed as an emergency contraceptive, with similar success in preventing pregnancy. However, vomiting occurred less frequently in women taking levonorgestrel alone. The Task Force on Postovulatory Methods of Fertility Regulation conducted a double-blind, randomized trial to evaluate the effectiveness of levonorgestrel compared with the Yuzpe regimen in preventing pregnancy.
Healthy women who reported regular menstrual cycles and had had only one act of unprotected intercourse within 72 hours of starting the regimen were eligible for the study. Women were excluded if they were breast-feeding, currently using hormonal contraception, had contraindications to the use of hormones for contraception or were uncertain about the date of their last menstrual cycle. Women included in the study provided medical and gynecologic histories and underwent physical examination and pregnancy testing at enrollment. Each participant was given two sets of two tablets. One group received levonorgestrel and the other the Yuzpe regimen. The regimens were identical in appearance and method of administration. The first dose was taken on enrollment at the clinic, and the second dose was taken 12 hours later at home. Each woman was also given one additional dose to be taken if she vomited within four hours of taking a dose. Participants kept diaries of side effects for one week following the second dose and recorded bleeding patterns and dates of any subsequent intercourse. All of the women were asked to return for follow-up assessment and pregnancy testing one week after the expected date of their next menstrual period.
Of the 1,955 women enrolled in the study, 976 received levonorgestrel alone and 979 received the Yuzpe regimen. Women in the levonorgestrel group had 11 (1.1 percent) documented pregnancies compared with 31 pregnancies (3.2 percent) in the Yuzpe group. Patient reports about their intercourse patterns indicated that 75 pregnancies would be expected in the levonorgestrel group and 72 pregnancies would be expected in the Yuzpe group if no emergency contraception were used. Levonorgestrel prevented pregnancy in 85 percent of women; the Yuzpe regimen prevented pregnancy in 57 percent of women. Nausea, vomiting, dizziness and fatigue were less frequent in the levonorgestrel group. Approximately 50 percent of women taking the Yuzpe regimen reported nausea and 18 percent reported vomiting compared with 23 percent reporting nausea and 5 percent reporting vomiting in the levonorgestrel group. The time to return of menstrual periods was similar in both of the groups.
The authors conclude that levonorgestrel was more effective and better tolerated than the Yuzpe regimen for emergency contraception. In addition, the authors believe that treatment is most effective when initiated as soon as possible after unprotected intercourse.
ANNE D. WALLING, M.D.
Task Force on Postovulatory Methods of Fertility Regulation. Randomised controlled trial of levonorgestrel versus the Yuzpe regimen of combined oral contraceptives for emergency contraception. Lancet August 8, 1998;352:428-33.
Intra-articular Corticosteroids for Chronic Arthritis in Children
Experience with intra-articular corticosteroid injections in children with arthritis is limited, primarily because of concern about infection and the potential for cartilage damage. However, reports in the literature seem to attest to its safety in children. Padeh and Passwell evaluated the efficacy of intra-articular corticosteroid injections in 71 children treated at a pediatric rheumatology clinic from 1992 through 1996.
The 47 girls and 24 boys in the study were six months to 18 years of age (mean age: 9.4 years). Sixty-one of the children fulfilled the American College of Rheumatology criteria for the diagnosis of juvenile rheumatoid arthritis. Six children had reactive arthritis, two had juvenile psoriatic arthritis, one had Crohn's disease and one had mixed connective tissue disease.
Most of the patients had not responded to six to eight weeks of therapy with an oral nonsteroidal anti-inflammatory agent, most commonly naproxen. They had objective signs of inflammatory arthritis, including joint swelling, effusion, warmth or decreased range of motion. Four patients had pauciarticular disease that either had not been treated or was misdiagnosed, and three children had developed joint deformities such as flexion contractures or leg-length discrepancies by the time they were referred.
Intra-articular injections were performed under local anesthesia with 2 percent lidocaine; sedation was used in children younger than six years and in children who had more than four joints injected at a time. Triamcinolone hexacetonide was administered in a dose of 10 to 40 mg, depending on the size of the joint. Patients were asked to limit themselves to nonweight-bearing activities for 24 hours after injection of the lower limbs and were seen one week later for a follow-up visit. A "good" response was defined as complete resolution of symptoms within one week and lasting for at least six months. A failed response was classified as a relapse of active synovitis within six months of the injection.
A total of 300 joint injections were performed, including injections in 124 knees, 71 ankles, 46 wrists, 12 tendon sheaths, 10 shoulders, seven elbows and 30 metacarpophalangeal and proximal interphalangeal joints.
Full remission that lasted more than six months occurred with 246 (82.0 percent) of the 300 injections. In 43 children (60.6 percent), it became possible to discontinue all oral medications, including nonsteroidal anti-inflammatory agents, sulfasalazine (four patients) and corticosteroids (three patients). Correction of joint contractions occurred in 42 children (55 joints), and Baker's cysts resolved completely in 11 patients. Inflammation recurred within six months in 54 (18.0 percent) instances.
There were no joint infections or other late effects such as joint instability, osteonecrosis or soft tissue atrophy during a follow-up that ranged from five to 48 months (average follow-up: 30 months). Two children had a postinjection flare-up that was apparent 24 hours after the injection but resolved within three to five days. Twenty children who had more than one injection of the same joint underwent radiographic evaluation, and no calcifications in or around the joints were visualized.
The authors conclude that intra-articular corticosteroid injections are safe and effective in children with inflammatory arthritis. The authors note that advantages of intra-articular corticosteroid injections include a rapid response, a low incidence of side effects and avoidance of long-term systemic therapy.
JEFFREY T. KIRCHNER, D.O.
Padeh S, Passwell JH. Intraarticular corticosteroid injection in the management of children with chronic arthritis. Arthritis Rheum July 1998;41:1210-4.
Efficacy of Tramadol vs. Hydrocodone-Acetaminophen
Tramadol hydrochloride is a centrally acting analgesic with minor opioid agonist properties that make it suitable for the treatment of acute pain. It also inhibits monoamine reuptake, which may make it suitable for the treatment of chronic pain. It has a lower potential for abuse than other opioids and no antiprostaglandin activity. Comparative studies have shown that tramadol hydrochloride has efficacy comparable to that of propoxyphene or pentazocine in the management of postoperative pain, and comparable to that of codeine after dental extraction. Turturro and associates conducted a randomized, double-blind, prospective study to evaluate the analgesic effect of oral tramadol hydrochloride versus oral hydrocodone with acetaminophen in patients with acute musculoskeletal pain after minor trauma.
Patients with suspected substance abuse, a history of hypersensitivity to any of the analgesics used in this study, or known pregnancy or lactation were excluded from the study. Study participants received an unlabeled capsule that contained either 100 mg of tramadol hydrochloride or 5 mg of hydrocodone with 500 mg of acetaminophen. Pain was evaluated at baseline and at 30-minute intervals up to 180 minutes using a visual analog scale. Mean pain scores were similar at baseline among the study participants but significantly lower in the group taking hydrocodone with acetaminophen from 30 minutes through the end of the study period. The difference in pain scores between the two groups became more pronounced with each interval that followed the initial analgesic dose.
The authors conclude that tramadol hydrochloride is less effective in the management of acute pain in patients with musculoskeletal pain following trauma than hydrocodone with acetaminophen. In addition, the manufacturer's reports of such side effects as dizziness within seven days of use, nausea, constipation, headache and somnolence may also cause problems in both short and long medication courses. The decreased analgesic effect, relatively frequent side effects and increased cost make tramadol hydrochloride a less desirable choice in the treatment of minor to moderate acute musculoskeletal pain.
RICHARD SADOVSKY, M.D.
Turturro MA, et al. Tramadol versus hydrocodone-acetaminophen in acute musculoskeletal pain: a randomized double-blind clinical trial. Ann Emerg Med August 1998;32:139-43.
Perinatal Risk of Asthma Medication During Pregnancy
Poorly controlled asthma is associated with adverse health outcomes in pregnant women and their infants, but little is known about the effect recent changes in prescribing practices (such as increased use of bronchodilator and anti-inflammatory agents) have had on perinatal outcomes. Alexander and colleagues conducted a retrospective cohort study to assess perinatal outcomes in women with asthma who used asthma medications during pregnancy.
More than 800 women with asthma who delivered between 1991 and 1992 in a Canadian hospital were identified. The women were classified into three groups based on the therapy they received for their asthma: 375 women who used no asthma medication, 303 women who used a beta agonist only and 139 women who used steroids, either alone or in combination with other medications. The pregnancy outcomes of these women were compared with those of 13,709 women without asthma who delivered at the same hospital during the same period. Women with asthma, particularly those who were not taking asthma medication, were more likely to be nulliparous and unmarried than women without asthma. Women with asthma were also more likely to smoke, with the highest rates of smoking (38 percent) reported among those who did not receive asthma treatment.
A significantly increased risk of antenatal and postpartum hemorrhage was found in women with asthma. Those treated with steroids had the highest risk. Fifteen percent of mothers in this group experienced hemorrhage, compared with 8 percent of mothers without asthma, 11.6 percent of mothers with asthma who were not receiving treatment and 10 percent of mothers with asthma who were treated with beta agonists. The risk of pregnancy-induced hypertension was also higher in women with asthma who were treated with steroids (18 percent) compared with 10.5 percent of women without asthma. The infants of women with asthma who were treated with steroids were also significantly more likely to develop hyperbilirubinemia. No statistically significant differences were found for low birth weight, respiratory distress or congenital abnormalities among the groups.
The authors postulate that the increased risk of hemorrhage in mothers with asthma could be related to coagulation changes associated with asthma and may be independent of medication and other factors. The increased risks of hypertension and hyperbilirubinemia could be related to asthma or to the use of steroids. They note with concern the clustering of risk factors in mothers with asthma who were not receiving asthma medication.
ANNE D. WALLING, M.D.
Alexander S, et al. Perinatal outcomes in women with asthma during pregnancy. Obstet Gynecol September 1998;92:435-40.
Is Documentation of Child Physical Abuse Improving?
For the past 10 years, the medical literature has increasingly provided information about physical and behavioral characteristics of child abuse. Thorough documentation of physical findings is critical because the medical record is a legal document. A 1980 study on trauma and child abuse stated that adequate documentation of child abuse should include a history and description of the injury, information about previous injuries, and a decision about whether the history and the injuries are compatible. Limbos and Berkowitz conducted a retrospective chart review of emergency department records from 1980 and 1995 to determine if increased awareness and training has improved documentation of suspected child abuse.
A county hospital's emergency department medical records from 1980 and 1995 were included in the study if they specified a discharge diagnosis of "child abuse" or "nonaccidental/intentional trauma." Children with diagnoses of neglect or sexual abuse were excluded from the review. Information about the patient's history, results of physical examination, diagnostic procedures, diagnosis and final disposition were obtained from each record.
Of the 25,500 children evaluated in the emergency department in 1980, 44 children met the diagnostic criteria for review. Of the 25,470 children evaluated in the emergency department in 1995, 31 children met the criteria, and 29 charts were available for review. Documentation did not differ significantly between the two groups. While the specific injury or reason for evaluation was documented consistently in both years, more than 75 percent of all charts lacked information about witnesses to the alleged abuse. Whether there was a history of previous injury was noted in 59 percent of the 1980 charts, but in only 45 percent of the 1995 charts. Developmental history was not included in any of the records. The location of physical injury was well documented in both years, but only about one third of the 1980 charts fully described these injuries, compared with more than one half of the charts in 1995.
Examination of the genitalia was performed only about one half of the time in either year. Records of laboratory studies were similar in both sets of records. However, skeletal surveys were performed in more than 80 percent of children in 1980, but in only 38 percent of children in 1995. This finding may be explained by the fact that these examinations are not generally recommended now except in children younger than two years. Charts from 1995 were more likely to include a discharge plan than were records from 1980.
The authors conclude that overall documentation of child abuse did not improve significantly between 1980 and 1995. Despite improvements in recording the discharge plan, omissions in important historical information and physical findings still exist. They suggest the use of structured forms and standardized checklists to improve documentation. Teaching rounds in the emergency department could provide a forum for residents and physicians to discuss errors, omissions and ambiguities in actual records. In addition, the authors suggest that a camera be available in every emergency department to improve documentation of alleged abuse.
GRACE BROOKE HUFFMAN, M.D.
Limbos MP, Berkowitz CD. Documentation of child physical abuse: how far have we come? Pediatrics July 1998; 102:53-8.
Effects of Walking During the First Stage of Labor
Walking during the first stage of labor is believed to enhance progression and decrease the pain of labor. However, little research has been performed to evaluate the effects of the woman's position during labor. Bloom and colleagues performed a randomized study to determine if walking during the first stage of labor has any impact on the duration of labor or obstetric outcome.
The study included 536 women who were assigned to a walking group and 531 women who were assigned to a usual-care group (labor in bed). The two groups were similar with respect to race, age, nulliparity and cervical dilatation at baseline. All of the patients had spontaneous onset of labor at 36 to 41 weeks' gestation. Uterine contractions were regularly occurring and the cervix was dilated 3 to 5 cm when patients entered the study. Fetal membranes could be intact or ruptured, and both multiparous and nulliparous patients were included. Women who had breech presentations or any known complications were excluded from the study.
The women in the walking group were allowed to walk as they desired; 116 (21.6 percent) of the 536 women in this group did not walk during the first stage of labor. Patients who walked returned to their beds if they needed analgesia or when the second stage of labor began.
Routine electronic fetal monitoring was not used in either group, although intermittent surveillance was used when clinically indicated. Ineffective labor was suspected if the cervix did not dilate progressively in the first two hours after admission. If the fetal membranes were intact, amniotomy was performed, and the progress of labor was evaluated two hours later. Augmentation with intravenous oxytocin was instituted if no further cervical dilatation occurred in two to three hours and hypotonic uterine contractions were documented by an internal pressure catheter.
A nurse attended each walking woman, recording the number of minutes spent walking. In addition, women from both groups wore a pedometer that recorded the number of steps taken during the first stage of labor.
The walking time was 56 ± 46 minutes and the number of steps taken was 553 ± 801 in the patients who walked during the first stage of labor. The 116 patients who were assigned to the walking group but chose not to walk had a significantly shorter first stage of labor than those who walked. The duration of the first stage of labor was 5.5 ± 3.6 hours in the group that chose not to walk, compared with a duration of 6.2 ± 3.5 hours in the group that walked. In the women assigned to the usual-care group, the duration of the first stage of labor was 6.1 ± 3.5 hours.
No significant differences between the walking and usual-care groups were found in any characteristic of labor, including the need for oxytocin, duration of the first and second stages of labor, need for analgesia or development of chorioamnionitis. There were also no differences in the frequency of episiotomy, use of forceps or need for cesarean section. Infant outcomes were similar in both groups.
The authors conclude that walking during the first stage of labor neither enhances nor impedes the normal progress of labor during the first stage. Because their study provides no objective evidence for or against walking during the first stage of labor, they believe it is reasonable to allow the patient to decide what she prefers to do.
JEFFREY T. KIRCHNER, D.O.
Bloom SL, et al. Lack of effect of walking on labor and delivery. N Engl J Med July 9, 1998;339:76-9, and Cefalo RC, Bowes WA Jr. Managing labor: never walk alone [Editorial]. N Engl J Med July 9, 1998;339:117-8.
EDITOR'S NOTE: The findings from this study should not discourage physicians from encouraging patients to walk during the first stage of labor. Rather, the findings suggest that walking during labor is not medically superior or inferior to labor in bed. In an accompanying editorial, Cefalo and Bowes note that since walking provides no measurable benefit but also causes no obvious harm, physicians should allow patients to have the last word as to whether they walk or stay in bed during the first stage of labor. The authors further note that the only intrapartum intervention truly shown to reduce the rate of cesarean section is the presence of one-on-one support, such as that given by a nurse, nurse-midwife or doula.
J.K.
"Tips from Other Journals" are written by the medical editors of American Family Physician.
Copyright © 1998 by the American Academy of Family Physicians.
This content is owned by the AAFP. A person viewing it online may make one printout of the material and may use that printout only for his or her personal, non-commercial reference. This material may not otherwise be downloaded, copied, printed, stored, transmitted or reproduced in any medium, whether now known or later invented, except as authorized in writing by the AAFP.
