Tips from Other Journals

Tips from Other Journals

Quinine for Treatment of Nocturnal Leg Cramps

The use of quinine for the treatment of nocturnal leg cramps (painful, involuntary muscle contractions that commonly occur in elderly patients) is common. Quinine has been shown to prevent the occurrence of these cramps, although it has not seemed to affect the severity or duration of cramps. Man-Son-Hing and colleagues had previously conducted a meta-analysis of published research on this topic; the current study includes data from unpublished studies collected by the U.S. Food and Drug Administration (FDA).

A literature search using the key words “quinine,” “muscle cramps” and “legs” yielded three well-designed published trials. Queries to the FDA and other authorities in the field yielded four unpublished reports that also met the criteria for this meta-analysis. To be accepted, the trials had to include general ambulatory patients and he randomized, double-blind and placebo-controlled, in a crossover design. The outcomes studied were reduction in number, severity and duration of nocturnal leg cramps.

Published trials showed larger reductions in the number of nocturnal leg cramps than did unpublished trials. Although combining the results of the unpublished trials with the published data reduced the apparent efficacy of quinine, the reduction in the number of leg cramps remained statistically significant. The relative risk reduction in published studies was 43 percent; in the unpublished studies it was 21 percent. As the treatment period lengthened, the relative risk reduction improved. The severity of leg cramps was also found to decrease significantly in patients treated with quinine. No data were available to determine if the duration of leg cramps was improved in patients taking quinine. These patients experienced more side effects, with tinnitus occurring significantly more often in treated patients than in those taking placebo.

The authors acknowledge the difficulties inherent in including unpublished data in a meta-analysis and conclude that adding the unpublished reports to the published reports shows that treatment of nocturnal leg cramps with quinine has some benefit, but perhaps not as much as was originally thought. They recommend that nonpharmacologic management (such as regular passive stretching of the leg muscles) be used as first-line treatment for this condition. If the patient experiences no improvement, a trial of quinine therapy (lasting at least four weeks) may be useful.


Man-Son-Hing M, et al. Quinine for nocturnal leg cramps. A meta-analysis including unpublished data. J Gen Intern Med September 1998; 13:600-6.

Chronic Hypertension and Adverse Neonatal Outcomes

Up to 5 percent of all pregnant women have underlying chronic hypertension, and rates are higher among women who are older, obese or black. Chronic hypertension increases the risk of preeclampsia and abruptio placentae, both of which may result in increased neonatal morbidity and mortality. However, data that identify specific risk factors for preeclampsia and other adverse pregnancy outcomes in women with chronic hypertension are limited. Sibai and colleagues evaluated data from a large multicenter prospective study of pregnant women with chronic hypertension.

The study group included women who were enrolled in a National Institute of Child Health and Human Development study that examined the potential benefits of low-dose aspirin therapy in the prevention of preeclampsia. All patients had chronic hypertension, defined as either systolic blood pressure of at least 140 mm Hg or diastolic blood pressure of at least 90 mm Hg on at least two occasions at least four hours apart before entry into the study or by 20 weeks of gestation. Preeclampsia was defined according to the baseline characteristics. In women with proteinuria at baseline, preeclampsia was defined as elevated levels of serum alanine transferase or elevated diastolic blood pressure, plus severe headaches, epigastric pain or increasing proteinuria. New-onset proteinuria was defined as results of at least + 1 on a routine urine dipstick test for protein. Abruptio placentae was diagnosed by clinical findings and placental examination. The women were divided into two groups: those who had proteinuria at baseline and those who did not. Neonatal outcomes that were evaluated included birth before 37 weeks’ gestation, birth weight below the 10th percentile for gestational age, admission to the neonatal intensive care unit and neonatal complications.

Of the 763 women included in the study, 381 received low-dose aspirin therapy, and 382 received placebo. Preeclampsia was diagnosed in 193 women (25 percent) and occurred with similar frequency in both groups. However, women with a prior history of preeclampsia, a diastolic blood pressure of 100 to 110 mm Hg early in the pregnancy or a history of hypertension for at least four years were more likely to have preeclampsia. Maternal age, race or urinary protein excretion at baseline were not associated with later development of preeclampsia. Abruptio placentae occurred in 11 women (1.5 percent). The frequency between groups was similar, except for women with preeclampsia, who were three times more likely to develop the condition. In addition, women with preeclampsia at baseline were significantly more likely to have preterm deliveries, to have more infants admitted to intensive care units and to have more neonatal morbidity and mortality. Similar neonatal outcomes were found in women with proteinuria at baseline. In addition, these women also had more infants who were small for gestational age.

The authors conclude that the incidence of preeclampsia is significantly higher in women who have a previous history of the condition, hypertension lasting at least four years or a diastolic blood pressure of at least 100 mm Hg early in the pregnancy. These findings are consistent with conclusions of earlier studies. Infants born to women with preeclampsia were found to have more complications, a finding that is also similar to previous studies. Of interest was the finding that neither black race nor advanced maternal age was a risk factor for superimposed preeclampsia. This finding differs from previously published studies. The presence of proteinuria early in the pregnancy appears to be a separate risk factor for adverse neonatal outcomes and underscores the importance of preconception counseling, screening and medical follow-up in patients with chronic hypertension.


Sibai BM, et al. Risk factors for preeclampsia, abruptio placentae, and adverse neonatal outcomes among women with chronic hypertension. N Engl J Med September 1998;339:667-71.

Outcome of Immunoprophylaxis in HBsAg-Positive Neonates

Mother-to-infant transmission of hepatitis B is more common among mothers positive for hepatitis B e antigen (HBeAg). Intrauterine infection, in which the infant is seropositive for hepatitis B surface antigen (HBsAg), remains a significant problem in the elimination of vertical hepatitis B transmission. Tang and associates examined the prevalence of HBsAg among infants born to HBeAg-positive, HBsAg-carrier mothers in Taiwan and the outcome of immunoprophylaxis in these infants.

The study included 665 infants born to HBeAg-positive, HBsAg-carrier mothers. Hepatitis B status was determined at birth and during long-term follow-up. All of the infants received hepatitis B immune globulin (HBIG) within 24 hours of birth and three or four subsequent doses of hepatitis B vaccine.

Sixteen (2.4 percent) of the 665 infants were seropositive for HBsAg within one hour of birth. All 16 infants remained HBsAg positive during the first six months of life. Follow-up in 12 of the 16 infants continued for 33 months to 12 years (mean follow-up: 8.1 years). In all but one child, the HBsAg-positive state (carrier state) persisted throughout the follow-up period. The child who showed seroconversion to hepatitis B surface antibody (anti-HBs) did so at four years of age. None of the children had abnormal serum alpha-feto-protein levels, or symptoms or signs of chronic liver disease.

The authors conclude that the current immunoprophylaxis strategy, including administration of HBIG within 24 hours after birth plus three or four doses of hepatitis B vaccine, fails to prevent chronic infection in infants who become infected in utero. The authors speculate that infants exposed to HBeAg and other hepatitis B antigens in utero may be more tolerant to the various antigens, including HBsAg, and unresponsive to hepatitis B vaccine. When immune tolerance wanes, seroconversion may occur.


Tang JR, et al. Hepatitis B surface antigenemia at birth: a long-term follow-up study J Pediatr September 1998; 133:374-7.

Metformin Use in Overweight Patients with Type 2 Diabetes A large prospective study in Great Britain reported that the use of sulfonylureas or insulin substantially reduced the risk of microvascular complications and, possibly, heart attacks in patients with type 2 diabetes. Metformin, a biguanide that decreases blood glucose concentrations by mechanisms different from those of sulfonylureas or insulin, theoretically could achieve glucose control without weight gain. The UK Prospective Diabetes Study Group examined the effect of intensive glucose control with metformin compared with that of other agents in overweight patients.

Between 1977 and 1991, 23 general practices enrolled 5,102 patients from 25 to 65 years of age who were newly diagnosed with type 2 diabetes. After three months of dietary therapy, the 4,209 eligible patients with fasting plasma glucose measurements above threshold levels were randomly assigned to continue diet therapy alone or to begin one of four drug therapies designed to achieve intensive glycemic control. Of the 1,704 patients who weighed over 120 percent of their ideal body weight, 411 were assigned to diet therapy, 342 to metformin therapy, 265 to chlorpropamide therapy, 277 to glibenclamide therapy (U.S. equivalent: glyburide) and 409 to insulin therapy.

Patients using metformin to control blood glucose were initially given 850 mg per day; the daffy dosage was titrated to a maximum of 2,550 mg. The dosage was reduced if nausea or diarrhea developed. If marked hyperglycemia developed during metformin therapy, glibenclamide was added; if the hyperglycemia persisted, treatment was changed to insulin. Episodes of hypoglycemia were recorded by patients throughout the study. Patients were seen every month for the first three months of treatment, then at least every three months thereafter, depending on the clinical situation. The median duration of follow-up was 10 years. Study end points included mortality related to diabetes and multiple markers of morbidity, such as retinopathy, nephropathy, peripheral vascular disease, amputation, stroke and cardiovascular disease.

Patients who received metformin had a 32 percent lower risk of developing any of the diabetes-related end points compared with patients using diet therapy alone. The patients using metformin also experienced a 42 percent reduction in death related to diabetes and a 36 percent reduction in all causes of mortality. When compared with patients receiving the other treatments, patients taking metformin showed significant reductions in any diabetes-related end point, all-cause mortality and stroke.

During the 10-year follow-up period, similar slight gains in body weight occurred in patients treated by diet and in patients receiving metformin; however, these gains were less than the increase in body weight that occurred in patients treated with sulfonylureas or insulin. In the group taking metformin, 8.3 percent of patients reported any episodes of hypoglycemia, and 0.6 percent experienced major episodes of hypoglycemia. The corresponding figures for patients in other treatment groups were: 7.9 percent/0.7 percent for diet therapy, 15.2 percent/1.2 percent for chlorpropamide, 20.5 percent/1.0 percent for glibenclamide and 25.5 percent/2.0 percent for insulin therapy. No deaths from lactic acidosis were reported during the study period.

The authors conclude that metformin therapy provided risk reduction in key end points for overweight patients with type 2 diabetes and was associated with less weight gain and fewer episodes of hypoglycemia than treatment with insulin or sulfonylureas. They advocate metformin as first-line pharmacologic therapy for overweight patients with type 2 diabetes.


UK Prospective Diabetes Study (UKPDS) Group. Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). Lancet September 12, 1998;352:854-65.

Is Counseling Beneficial in the Prevention of HIV and STDs?

The prevention of human immunodeficiency virus (HIV) infection is important since treatment, although improving, is still far from effecting a cure. Kamb and colleagues conducted this multicenter, randomized, controlled trial to determine whether counseling can prevent behaviors associated with an increased risk of contracting HIV and other sexually transmitted diseases (STDs).

Patients were recruited from inner-city STD clinics. Eligible patients were at least 14 years of age and were not HIV positive. Men who described themselves as homosexual or bisexual were excluded from the study. Patients were randomly assigned to one of four intervention groups: (1) enhanced interactive counseling, consisting of four counseling sessions (three of which were 60 minutes long) within four weeks of enrollment in the trial; (2) a brief interactive counseling intervention in two 20-minute sessions; (3) a noninteractive, didactic approach that stressed prevention in two five-minute sessions with follow-up appointments; and (4) a noninteractive didactic approach alone. The first three groups had scheduled follow-up appointments every three months for one year after enrollment.

Study subjects received free condoms and a relatively small financial incentive (no more than $45) for their participation. The study included 3,269 men and 2,489 women. The main outcomes observed by the authors were the participants’ self-reported use of condoms and the occurrence of STDs in patients during the follow-up period. The assigned intervention sessions were completed by 82 percent of the participants.

All intervention groups reported improved rates of “no unprotected vaginal sex” and increased use of condoms at three and six months compared with baseline. Participants in the one-on-one intervention groups were significantly less likely to have unprotected sex than were patients in the didactic groups. By the nine- and 12-month follow-up visits, use of condoms in all groups was still more frequent than at baseline, but there were no significant differences among the groups. Patients in the two interactive intervention groups were less likely to develop STDs over the course of the study than were patients in the didactic groups. When the patients in the didactic group with follow-up were compared with the patients in the didactic group without follow-up one year after enrollment, it was discovered that the patients in the former group were less likely to engage in high-risk behaviors (such as unprotected vaginal sex).

The authors conclude that the incidence of STDs decreased by about 30 percent in the first six months of follow-up after interactive HIV/STD counseling. This figure dropped to about 20 percent after one year of follow-up but still shows the value of patient-centered, tailored counseling. The counselors in the trial typically did not have advanced degrees or long experience in interactive counseling; rather, they used personalized risk reduction plans to increase condom use and prevent new cases of STDs.


Kamb ML, et al. Efficacy of risk-reduction counseling to prevent human immunodeficiency virus and sexually transmitted diseases. A randomized controlled trial. JAMA October 7, 1998;280:1161-7.

Practical Management of Breast Fibroadenomas

Fibroadenomas occur in about 10 percent of women seen in breast clinics, but account for about one half of all breast biopsies. These benign lesions occur more often in women of higher socioeconomic status and in women who are darker-skinned. The risk of developing fibroadenomas is not related to age of menarche or menopause, or to use of hormonal therapy. Conversely, higher body mass index, greater number of full-term pregnancies, higher intake of vitamin C and cigarette smoking are associated with a decreased risk of these lesions. Greenberg and associates present a literature-based approach to fibroadenomas and propose practical algorithms for their management.

Breast fibroadenomas are usually detected incidentally during clinical or self-examinations and are most often located in the upper outer quadrants. The majority are smooth, mobile, nontender and rubbery. Some women have multiple fibroadenomas; these patients usually have a family history of fibroadenomas. Rarely, adolescent girls may develop lesions larger than 5 cm, called juvenile (or giant) fibroadenomas. Although much larger than typical fibroadenomas, these giant fibroadenomas are also benign and do not become malignant. Sonography, often used to diagnose fibroadenomas, reveals a round or oval solid mass with a smooth contour and weak internal echoes. Sonography is able to distinguish between solid and cystic lesions but cannot reliably distinguish fibroadenoma from breast cancer. In one study, about 18 percent of fibroadenomas could not be seen in an ultrasound study of the breast. In a mammogram, a fibroadenoma appears as a soft, homogenous, circumscribed nodule, sometimes with inner coarse calcifications. Fine-needle aspiration reveals characteristic cytologic features and can be used to improve clinical diagnosis of fibroadenoma. Fine-needle aspiration rarely gives a false-positive result, even though it may confuse fibroadenomas with other benign lesions.

Depending on the patient’s age, family history and history of previous biopsies, conservative treatment of fibroadenomas may or may not be warranted. A woman under 25 years of age with fibroadenomas is unlikely to have a “missed” malignancy if she has had a clinical examination, negative sonography and negative fine-needle aspiration. (The risk of a missed malignancy in this situation has been estimated to be one in 229 cases to one in 700 cases.) It is not until age 35 that the risk increases somewhat. Multiple fibroadenomas should be excised, with care taken to avoid scarring or ductal damage, which may occur when one incision is used. Giant fibroadenomas generally shrink after hormonal balance is achieved (e.g., after cessation of lactation), so conservative management is preferred, with excision warranted after the lesions become smaller.

The authors conclude that, in women younger than 35 years, conservative management of fibroadenomas is recommended, with a follow-up every six to 12 months until complete regression. Fibroadenomas that don’t regress or that remain unchanged by age 35 should be excised. Women older than 35 years should have mammograms in addition to the above-mentioned diagnostic modalities, with follow-up after six to 12 months. If the fibroadenoma has not resolved after 12 months, excision is recommended. Even when conservative management is reasonable, patients often prefer excisional biopsy of persistent fibroadenomas. The authors advise excisional biopsy of any mass for which a diagnosis of fibroadenoma is not clear-cut; however, if, after a clinical examination, ultrasound examination and fine-needle aspiration, a diagnosis of multiple fibroadenomas can be confidently made, conservative treatment is warranted.


Greenberg, et al. Management of breast fibroadenomas. J Gen Intern Med September 1998; 13:640-5.

EDITOR’S NOTE: When mammography is going to be used to assist in the diagnosis of fibroadenomas, it should be performed before fine-needle aspiration, so that subsequent hematoma formation does not distort the mammographic picture. It should also be emphasized that when the nature of a lesion is not definite after ultrasonography, mammography and fine-needle aspiration, the lesion should be excised.-G.B.H.

Differential Diagnosis of Cyclic Vomiting in Children

Cyclic vomiting is defined as episodes of vomiting interspersed with periods of wellness. Patients with cyclic vomiting have at least four episodes of vomiting per hour, but no more than two episodes per week. After testing, no cause of the vomiting is found. Previously, the vomiting has been attributed to migraine (so-called abdominal migraine), hypothalamic dysfunction or problematic parent-child relationships. Li and colleagues conducted a chart review of children who presented with an episodic pattern of vomiting to determine which disorders should be included in the differential diagnosis for a work-up of a child with cyclic vomiting.

Children who had an episodic pattern of vomiting with at least three discrete episodes were included in the study. Results of laboratory tests, radiographs and endoscopic biopsies, and clinical follow-up were reviewed. A structured telephone interview elicited a history of the pattern of vomiting, associated symptoms and possible trigger factors.

The mean occurrence of vomiting was 6.3 episodes per hour and 1.6 episodes per month. In slightly more than one half of the children (53 percent), a single diagnosis was determined to be the probable cause of the vomiting. Of these children, 88 percent were diagnosed with idiopathic cyclic vomiting syndrome, 7 percent were diagnosed with gastrointestinal conditions and 5 percent were diagnosed with extraintestinal problems.

When the treatments of children with multiple diagnoses were reviewed, 18 of 83 children responded, at least partially, to medication given for a diagnosed problem that was thought to have caused the vomiting (e.g., patients with vomiting and esophagitis partially responded to therapy with [H.sub.2] antagonists and cisapride). Twenty-six confirmed surgical disorders, such as small bowel malrotations, cholelithiasis and intracranial neoplasms, were believed to be associated with the vomiting. Various diagnostic tests, including blood counts, glucose levels, electrolyte studies and blood urea nitrogen levels, were performed in most patients during vomiting episodes. Forty-three percent of endoscopies yielded positive results, as did 38 percent of radiographs of the sinus, 28 percent of radiographs of the small bowel and 28 percent of computed tomographic scans or magnetic resonance images of the brain. As would be expected, patients who vomited more or had more frequent episodes of vomiting were tested more often than patients with less severe episodes.

The authors conclude that a rational laboratory evaluation of cyclic vomiting may be initiated based on the results of this study. The tests with the highest yield of positive results in this study were endoscopy and radiography of the sinus and small bowels. More specialized testing, such as determination of blood glucose, electrolytes, liver enzymes, pancreatic enzymes and ammonia levels, should be performed during the vomiting episode, when possible, to maximize the chance of detecting an intermittent or heterozygous disorder. Finally, surgical disorders are best detected by radiographs of the small bowel, abdominal imaging and cranial imaging.


Li BU, et al. Heterogeneity of diagnoses presenting as cyclic vomiting. Pediatrics September 1998; 102:583-7.

Oral vs. Vaginal Misoprostol for the Induction of Labor

Misoprostol, a synthetic prostaglandin [E.sub.1] analog, can initiate uterine contractions and has been reported to effectively induce labor. Some studies have found that vaginally administered misoprostol resulted in a shorter time to delivery than orally administered misoprostol but was associated with tachysystole and hyperstimulation of the uterus, which theoretically could lead to increased risk of operative delivery and neonatal asphyxia. Bennett and colleagues compared the effectiveness and incidence of adverse effects of misoprostol administered orally with misoprostol given vaginally in the induction of labor in women who were of at least 37 weeks’ gestation.

Data were compared from 206 Canadian women who met the criteria for safe induction of labor for anticipated vaginal delivery. After screening and giving informed consent, patients were randomly assigned to receive misoprostol either orally or vaginally. To ensure the double-blind nature of the study, each patient received either active oral misoprostol (50 mg) plus vaginal placebo or active vaginal misoprostol (50 mg) plus oral placebo every four hours until the occurrence of one of the following: at least three contractions every 10 minutes, spontaneous rupture of the membranes or delivery, or a concern about fetal heart rate or other complications. The management of labor and delivery was determined by attending physicians, and each patient was monitored throughout the process. Patient satisfaction was assessed by questionnaire following delivery.

The mean time to delivery was significantly reduced in the 102 women treated vaginally (846 minutes) compared with the 104 treated orally (1,072 minutes). The two groups did not differ in analgesia, epidural anesthetic or oxytocin use. The two groups also had similar patterns of delivery with no difference in the rate of cesarean section or assisted delivery. Vaginal misoprostol was associated with increased rates of tachysystole and hyperstimulation, but the differences did not reach statistical significance. Neonatal outcomes were similar in both groups. The median Apgar scores at one and five minutes were identical in the two groups; however, more infants in the group whose mothers received vaginal misoprostol had an Apgar score of less than seven at one minute. Patient-reported data also showed no difference in the incidence of gastrointestinal upset or in general satisfaction.

The authors conclude that both oral and vaginally administered of misoprostol are safe and effective for induction of labor. Vaginal administration is associated with more rapid delivery, greater uterine activity and more frequent fetal heart rate graph abnormalities. This finding may reflect uncertainties about the optimal dosage of vaginal misoprostol. Until the optimal dosing interval for vaginal use is determined, the preferred route of misoprostol administration might be oral.


Bennett KA, et al. A masked randomized comparison of oral and vaginal administration of misoprostol for labor induction. Obstet Gynecol October 1998; 92:481-6.

Xylitol for Prevention of Acute Otitis Media in Children

Because recurrent episodes of acute otitis media can lead to long-term sequelae such as learning disabilities, prevention of this condition would be more effective than separate treatment of each episode. Prophylaxis with antibiotics is often effective but can cause the development of resistant bacteria. The sweetening agent xylitol has been shown to prevent dental caries by inhibiting growth of Streptococcus mutans and has also been shown to inhibit growth of Streptococcus pneumoniae. Uhari and colleagues conducted a three-month randomized controlled trial to compare the effectiveness of xylitol syrup, chewing gum and lozenges in the prevention of acute otitis media in children.

Healthy children were recruited from 34 day care centers in Finland. History and risk factors for acute otitis media were obtained from the parents of each child. Children were not admitted to the study until all ear effusions were cleared, as determined by tympanometry and pneumatic otoscopy. A total of 857 children were screened and randomized to receive chewing gum, lozenges or syrup on the basis of their ability to chew gum. Each preparation was taken five times per day. Any child who developed symptoms of respiratory infection was examined by a nurse. Tympanometry was performed within three days of onset of symptoms. If the tympanogram was normal, the child was re-examined weekly. If results were abnormal, pneumatic otoscopy was performed. Acute otitis media was diagnosed if the child had the following: abnormal findings on tympanometry, middle ear effusion, symptoms of acute respiratory infection and signs of tympanic membrane inflammation. Children with acute otitis media were treated with antibiotics for seven days and reexamined weekly until the end of the study period.

At least one occurrence of acute otitis media was noted in 49 children (28 percent) in the control group receiving gum, 29 children (16 percent) in the group receiving xylitol gum, 39 children (22 percent) in the group receiving xylitol lozenges, 68 children (41 percent) in the control group receiving syrup find 46 children (29 percent) in the group receiving xylitol syrup. Some children had more than one episode of acute otitis media; the group receiving xylitol syrup had approximately 30 percent fewer episodes of acute otitis media than their counterparts who received control syrup. This difference was statistically significant. There was a decrease of approximately 40 percent in episodes of acute otitis media in those who received xylitol gum compared with those who received control gum. The patients receiving xylitol syrup and xylitol gum were given significantly fewer antibiotics than those in the control groups. The most common side effect was abdominal discomfort.

The authors conclude that xylitol chewing gum and xylitol syrup were associated with a significant reduction in the occurrence of acute otitis media and antibiotic use in children in day care centers. In a related editorial, Wright is cautiously optimistic about an agent that can decrease the occurrence of acute otitis media and the use of antibiotics. However, as he points out, “a biologically plausible mechanism of action” is missing, and the adverse effects of using such a substance throughout early childhood are unknown. In another editorial, Mitchell notes that xylitol is not readily available in the United States. This preliminary study is promising; however, further study is needed to determine the full range of adverse effects, to find the optimal dosage of xylitol and to determine the duration of xylitol’s prophylactic benefits.


Uhari M, et al. A novel use of xylitol sugar in preventing acute otitis media. Pediatrics October 1998;102:879-84, Wright PE Xylitol sugar and acute otitis media [Editorial]. Pediatrics October 1998;102:971-2, and Mitchell AA. Xylitol prophylaxis for acute otitis media: tout de suite? [Editorial]. Pediatrics October 1998; 102:974-5.

CT to Exclude Subarachnoid Hemorrhage in Severe Headaches

Previous studies suggest that approximately 10 percent of subarachnoid hemorrhages are missed by noncontrast computed tomography (CT). The Stroke Council of the American Heart Association strongly recommends that lumbar puncture be performed if CT scans are negative in patients who present with suspected subarachnoid hemorrhage. Although cerebrospinal fluid analysis may provide important information, the discomfort, time and expense of lumbar puncture may inhibit its use. Morgenstern and associates assessed the adequacy of CT imaging in excluding subarachnoid hemorrhage and the value of various cerebrospinal fluid measurements in the diagnosis of subarachnoid hemorrhage.

This prospective study included 107 patients who presented with “the worst headache of their life.” All of the patients underwent CT scan of the head, and evidence of subarachnoid hemorrhage was noted in 18 patients (16.8 percent). If CT scans were negative, lumbar puncture was performed. Lumbar puncture was performed in 79 of the 89 patients in whom CT scanning did not show evidence of hemorrhage. It was not performed in 10 patients who refused the procedure. Subarachnoid blood was found in two of the 79 patients (2.5 percent) who underwent lumbar puncture.

Patients were considered to have subarachnoid hemorrhage on the basis of spinal fluid results if the sample contained more than 1,000 red blood cells and no decrement in the count of more than 25 percent was noted from the first to the last tubes. In addition, analysis of cerebral spinal fluid had to meet the following criterion: positive results on spectrophotometry, visual xanthochromia or D-dimer assay.

The authors conclude that modern CT imaging will exclude 97.5 percent of subarachnoid hemorrhages in patients who present with the worst headache they have ever experienced. Subarachnoid hemorrhage will be detected by analysis of cerebrospinal fluid in approximately 2 percent of patients who present with “worst headache” symptoms and have negative findings on CT scan. Because of the small sample in this study, the calculated 95 percent confidence interval for missing subarachnoid hemorrhage on CT scan was 03 to 8.8 percent. This number is lower than that reported with earlier-generation CT scanners. The authors recommend that patients with the worst headache of their lives undergo lumbar puncture if noncontrast CT scanning fails to show evidence of subarachnoid hemorrhage.


Morgenstern LB, et al. Worst headache and subarachnoid hemorrhage: prospective, modern computed tornography and spinal fluid analysis. Ann Ernerg Med September 1998;32:297-304.

“Tips from Other Journals” are written by the medical editors of American Family Physician.

COPYRIGHT 1999 American Academy of Family Physicians

COPYRIGHT 2000 Gale Group