Comparing methods of clinical measurement: reporting standards for bland and altman analysis

In this era of medical technology assessment and evidence-based medicine, evaluating new methods to measure physiologic variables is facilitated by standardization of reporting results. It has been proposed that assessing repeatability be followed by assessing agreement with an established technique. If the "limits of agreement" (mean bias +/- 2SD) are not clinically important, then one could use two measurements interchangeably. Generalizability to larger populations is facilitated by reporting confidence intervals. We identified 44 studies that compared methods of clinical measurement published during 1996 to 1998 in seven anesthesia journals. Although 42 of 44 (95.4%) used the limits of agreement methodology for analysis, several inadequacies and inconsistencies in reporting the results were noted. Limits of agreement were defined a priori in 7.1%, repeatability was evaluated in 21.4%, and relationship (pattern) between difference and average was evaluated in 7.1%. Only one of the articles reported confidence intervals. A computer macro for the Minitab statistical package (State College, PA) is described to facilitate reporting of Bland and Altman analysis with confidence intervals. We propose standardization of nomenclature in clinical measurement comparison studies.

IMPLICATIONS

A literature review of anesthesia journals revealed several inadequacies and inconsistencies in statistical reports of results of comparison studies with regard to interchangeability of measurement methods. We encourage journal editors to evaluate submissions on this subject carefully to ensure that their readers can draw valid conclusions about the value of new technologies.

Reaction of gadolinium chelates with endogenously available ions

The extent of reaction of 153Gd-radiolabeled Gd(L) chelates with 25 mM CO23- (25 mF), PO34-, Zn2+ and Cu2+ at pH 7 was determined for L = EDTA, DTPA, DOTA, HP-DO3A, and DO3A. Gd(EDTA)- and Gd(DTPA)2- reacted (greater than 20% in 10 min) with Cu2+ and Zn2+ in the presence of PO34-. These double replacement reactions yielded precipitated GdPO4 and chelated Cu(L). Gd(HP-DO3A), Gd(DO3A) and Gd(DOTA)- were inert to reaction with all four ions at room temperature (less than or equal to 1% reaction detected). The thermodynamic binding constants of the ligands for Gd3+ and Cu2+ were found to be equal (10(20) M-1) for DO3A, while DOTA and HP-DO3A favored Gd3+ over Cu2+ by greater than or equal to 10(2) M-1. The low order of reactivity of Gd(DOTA)- and Gd(HP-DO3A) was anticipated by the binding constants, but the lack of reactivity of Gd(DO3A) is attributed to kinetic inertia. This latter property, desirable in MRI contrast agents, is promoted by the conformational stability of the tetraazacyclododecane macrocycle, which forms the backbone of the ligand. It is concluded that this class of chelates is exceptionally inert in solutions of endogenously available ions, and that thermodynamics alone is an insufficient predictor of the reactivity of the highly inert Gd complexes based on the tetraazamacrocycle.

Assessing the risk of venous thromboembolic events in women taking progestin-only contraception: a meta-analysis

OBJECTIVES

To evaluate the risk of venous thromboembolic events associated with the use of progestin-only contraception and whether that risk differs with the mode of drug delivery (oral, intrauterine, or depot injection).

DESIGN

Systematic review and meta-analysis of randomised controlled trials and observational studies.

DATA SOURCES

Pubmed, Embase, Cochrane Library, and reference lists of relevant reviews.

STUDY SELECTION

Randomised controlled trials and case-control, cohort, and cross sectional studies with venous thromboembolic outcome for progestin-only contraception reported relative to a non-hormone comparator group.

DATA EXTRACTION

Data were extracted by two independent investigators, and consensus for inclusion was reached after assessment by additional investigators.

RESULTS

Among the 2022 unique references identified by all searches, eight observational studies fulfilled inclusion criteria. A total of 147 women across all studies were diagnosed with a venous thromboembolic event while taking progestin-only contraception, and the summary measure for the adjusted relative risk of a venous thromboembolic episode for users versus non-users of a progestin-only contraceptive was, based on the random effects model, 1.03 (95% CI 0.76 to 1.39). Subgroup analysis confirmed there was no association between venous thromboembolic risk and progestin-only pills (relative risk 0.90 (0.57 to 1.45)) or a progestin intrauterine device (0.61 (0.24 to 1.53)). The relative risk of a venous thromboembolic event for users of an injectable progestin versus non-users was 2.67 (1.29 to 5.53).

CONCLUSIONS

Published data assessing the risk of venous thromboembolism in women prescribed progestin-only contraception are limited. In this meta-analysis of eight observational studies, the use of progestin-only contraception was not associated with an increased risk of venous thromboembolism compared with non-users of hormonal contraception. The potential association between injectable progestins and thrombosis requires further study.

A proposal to use confidence intervals for visual analog scale data for pain measurement to determine clinical significance

Visual analog scales (VAS) ranging from 0 cm (no pain) to 10 cm (worst imaginable pain) are used widely for pain measurement, but various investigators have not treated these data consistently. Conventional statistical tests of such data, although evaluating the "statistical significance" may obscure the clinical value of a treatment. On the other hand, confidence intervals (CIs) can illuminate both statistical and clinical importance. CIs give a range of values based on the observed data which contain, with a specified probability, a true but unknown variable typifying a population. We reviewed 112 articles published recently in anesthesia journals for statistical reporting of VAS data. Of the 112 articles, only two used CIs to report mean pain scores and one used CIs to report differences in median pain scores between the study groups. Only two articles presented 95% CI for the mean pain scores graphically. Analgesic techniques that produce VAS values in the range of 0-3 have been reported to represent adequate analgesia. A graphical method using CIs is proposed that allows ready interpretation of VAS data. With this approach, one evaluates whether the 95% CI for the mean pain score in a group during a particular period lies entirely within the zone defined as "analgesic success" (0-3). Such an analysis allows a visual assessment of whether a particular technique would produce clinically important effects in the population at large. This approach seems to provide more information than the use of conventional hypothesis testing in the interpretation of VAS data for pain measurement.

Premedication with oral and transdermal clonidine provides safe and efficacious postoperative sympatholysis

We studied 61 patients undergoing elective major non-cardiac surgery in a randomized, double-blind, placebo-control clinical trial to test the hypothesis that the addition of clonidine to a standardized general anesthetic could safely provide postoperative sympatholysis for patients with known or suspected coronary artery disease. Patients were allocated randomly to receive either placebo (n = 31) or clonidine (n = 30). The treatment group received premedication with a transdermal clonidine system (0.2 mg/d) the night prior to surgery, which was left in place for 72 h, and 0.3 mg oral clonidine 60-90 min before surgery. Clonidine reduced enflurane requirements, intraoperative tachycardia, and myocardial ischemia (1/28 clonidine patients vs 5/24 placebo, P = 0.05). However, clonidine decreased heart rates only during the first five postoperative hours; the incidence of postoperative myocardial ischemia (6/28 clonidine vs 5/26 placebo) did not differ between the two groups. Patients who experienced postoperative myocardial ischemia tended to have higher heart rates after surgery. Clonidine significantly reduced the plasma levels of epinephrine (P = 0.009) and norepinephrine (P = 0.026) measured on the first postoperative morning. There were no differences in the need for intravenous fluid therapy or antihypertensive therapy after surgery. The number of hours spent in an intensive care setting and the number of days spent in hospital were not different between the two groups. These results suggest that larger doses of clonidine should be investigated for their ability to decrease postoperative tachycardia and myocardial ischemia.

Genetic modification of prenatal lethality and dilated cardiomyopathy in Mn superoxide dismutase mutant mice

Mn superoxide dismutase (MnSOD), a mitochondrial antioxidant enzyme, has been shown to be essential for animal survival. MnSOD mutant mice (Sod2-/- mice) on the CD1 background develop severe dilated cardiomyopathy and usually die within 10 d after birth. To characterize better the phenotype and understand the mechanism of superoxide-mediated tissue damage in Sod2-/- mice, congenic Sod2-/- mice on inbred backgrounds were generated to ensure genetic homogeneity. When generated on a C57BL/6J background (B6<Sod2-/->), more than half of the fetuses develop severe dilated cardiomyopathy by embryonic day 15 and die in the uterus. Those that survive to term usually die within 24 h. In contrast, Sod2-/- mice on DBA/2J (D2<Sod2-/->) and B6D2F1 (B6D2F1<Sod2-/->) backgrounds develop normally throughout gestation and do not develop dilated cardiomyopathy. However, the D2<Sod2-/-> mice do develop a severe metabolic acidosis and survive for only up to 12 d after birth. B6D2F1<Sod2-/->) mice have a milder form of metabolic acidosis and can survive for up to 3 weeks. The marked difference in lifespans and the development of dilated cardiomyopathy in the B6 but not the D2 or B6D2F1 backgrounds indicate the possible existence of genetic modifiers that provide protection to the developing hearts in the absence of MnSOD.

Relative effectiveness of four preoperative tests for predicting adverse cardiac outcomes after vascular surgery: a meta-analysis

Various noninvasive tests have been proposed to stratify perioperative cardiovascular risk, including dipyridamole thallium scintigraphy (DTS), ejection fraction estimation by radionuclide ventriculography (RNV), ischemia monitoring by ambulatory electrocardiography (AECG), and dobutamine stress echocardiography (DSE). Which of these tests is most effective for predicting adverse perioperative cardiac outcome? To answer this question, and also to stimulate future studies, we evaluated 56 studies examining one or more of the four tests. We conducted meta-analysis on 20 studies that met the inclusion criteria. Outcome measures evaluated were cardiac death or myocardial infarction occurring during hospital stay or within 1 mo after surgery. Relative risk (RR), which is the probability of adverse cardiac outcome when the test is positive divided by the probability of adverse outcome when the test is negative, was used to combine evidence from different studies. An empirical Bayes procedure with a normal-normal hierarchic model was then used to obtain a meta-analytic confidence interval for the overall median of the relative risks. The between-study variance was estimated using the method of moments approach described by DerSimonian and Laird (Controlled Clin Trials 1986;7:177-88). Combined (median) RR [95% confidence interval (CI)] and the number of studies included in our meta-analysis for different evaluative tests were as follows: DTS 4.6 (2.1-10.4) (n = 6); RNV 3.7 (1.6-8.3) (n = 5); AECG 2.7 (1.4-5.1) (n = 6), and DSE 6.2 (1.7-22.8) (n = 3). We conclude that while DTS, RNV, AECG, and DSE are effective (the 95% CIs are greater than 1.0) in predicting the cardiac outcome after vascular surgery, the data are not definitive in determining the optimal test (95% CIs for RR overlap). Future studies should include DSE, as this test shows great promise for predicting adverse cardiac events after vascular surgery.

Low molecular weight heparin to achieve live birth following unexplained pregnancy loss: a systematic review

BACKGROUND

The management of recurrent pregnancy loss is uncertain. Some cohort studies have identified an association between inherited thrombophilias and recurrent or late non-recurrent pregnancy loss, which has prompted investigators to evaluate the benefit of low molecular weight heparin (LWMH) to achieve live birth. A similar benefit for LMWH has also been proposed independent of thrombophilia status.

OBJECTIVE AND METHODS

We conducted a systematic review of randomized controlled trials to assess the benefit of LMWH in achieving live birth for women with a history of recurrent or late non-recurrent pregnancy loss in the absence of antiphospholipid antibodies.

RESULTS

For the five studies that satisfied the eligibility criteria, the risk ratio of live birth for women with a history of pregnancy loss treated with LWMH compared with control ranged from 0.95 to 3.00. There was considerable heterogeneity among studies in terms of treatment effect (Q-value was 41.7, P=0.000, and I2=90.4%) independent of thrombophilia status. There was also a wide variation among all studies in terms of definition of early or late pregnancy loss, thrombophilic risk factors, and number of prior pregnancy losses.

CONCLUSION

There is a trend for increased live births when using LWMH for the prevention of recurrent pregnancy loss. Currently, there is insufficient evidence to support the routine use of LWMH to improve pregnancy outcomes in women with a history of pregnancy loss. Not only are additional studies necessary but standardized criteria for trials evaluating the benefit of an intervention in recurrent pregnancy loss should be established.

Lack of effect of food on the oral bioavailability of irbesartan in healthy male volunteers

This study was conducted to evaluate the effects of a high-fat meal on the oral bioavailability of an 300-mg irbesartan tablet in healthy male volunteers. Sixteen healthy young male volunteers participated in this single-center, open-label, single-dose, crossover study. Each volunteer received a single 300-mg irbesartan tablet under fasted conditions and 5 minutes after a high-fat breakfast, with administrations separated by a 7-day washout period. Serial blood samples were collected over a 72-hour period, and plasma samples were analyzed for irbesartan using a validated high-performance liquid chromatography/fluorescence procedure. Food had no statistically significant effects on the peak concentration (Cmax) and area under the concentration-time curve (AUC) of irbesartan. The presence of food was associated with a slightly prolonged time to maximum concentration (tmax) and half-life (t1/2), but the differences were not statistically significant. The results of this study indicate that food does not affect the bioavailability of irbesartan. Thus, irbesartan can be administered without regard to meals.

Specific radioimmunoassays for the measurement of stavudine in human plasma and urine

Sensitive and specific radioimmunoassays (RIAs) have been developed and validated for the determination of stavudine, a nucleoside analog possessing anti-human immunodeficiency virus (HIV) activity, in human plasma and urine. The hemisuccinate of stavudine was conjugated with histamine and radioiodinated to yield the radiotracer. Antisera were produced by injecting the immunogen, stavudine-hemisuccinate-bovine thyroglobulin, into rabbits. The antisera exhibited high specificity for stavudine as the structurally related analogs and other anti-HIV agents did not interfere in the assays. The methods could reliably quantitate stavudine in plasma from 2.5-100 ng ml-1 and in urine from 5.0-1000 ng ml-1 (after 2.5-fold dilution) with good accuracy and precision. The lower limits of quantitation were 2.5 ng ml-1 in human plasma and 5.0 ng ml-1 in urine (after 2.5-fold dilution). The RIA methods were applied to the analysis of stavudine in plasma and urine obtained from HIV-infected patients receiving the drug in clinical trials.

Visual evoked potentials and visual acuity after transurethral resection of the prostate

Changes in visual evoked potentials, visual acuity, blood ammonia levels and serum electrolytes (Na+ and K+) after transurethral resection of the prostate using glycine as an irrigating fluid performed under subarachnoid block were studied in 12 patients, in the pre-operative and immediate postoperative periods. Visual evoked potentials (p100 latency), recorded by shift of a checkerboard pattern, increased significantly from a pre-operative value of mean (SEM) 101.18 (1.63) msec in the right eye, and 102.5 (1.47) msec in the left eye to 108.91 (1.8) msec (p less than 0.01) and 108.08 (2.53) msec (p less than 0.01) respectively in the postoperative phase. There were no changes in visual acuity as assessed by a Snellen's chart, blood ammonia levels and serum electrolyte concentrations. The amount of glycine used intra-operatively for irrigation ranged from 3 to 31 litres.

A specific radioimmunoassay for the measurement of gadoteridol, a contrast agent for magnetic resonance imaging in biological fluids

Gadoteridol, a nonionic gadolinium chelate, is currently being evaluated for contrast-enhanced magnetic resonance imaging. A radioimmunoassay (RIA) has been developed for the measurement of gadoteridol in biological fluids. The RIA has a range of 0 to 25 micrograms/mL and has the sensitivity to detect 0.05 microgram/mL of gadoteridol. Satisfactory zero binding and sensitivity were obtained after an overnight incubation at 4 degrees C. Separation of the antibody-bound and free radiolabel was achieved with 12.5% polyethylene glycol. A quantitative recovery of the exogenous analyte was obtained at all concentrations of gadoteridol tested. Linearity in both serum and urine was satisfactory. Intraassay coefficients of variation were 6.4 and 2.8% for the low and medium controls, respectively. Interassay coefficients of variation were 5.4, 3.8, and 12.2% for the low, medium, and high controls, respectively. Cross reactivities of the ligand 5 and the calcium salt 6 were 37 and 29%, respectively. Clinical samples from the ascending dosage studies were analyzed by the gadoteridol RIA. The results obtained from the serum specimens demonstrated an excellent linear proportionality between drug concentration in blood and administered dosage of gadoteridol. Cumulative urinary excretion data showed that 94% of the drug was excreted in the urine within 24 h.

Radioimmunoassay for hydroxyphosphinyl-3-hydroxybutanoic acid (SQ 33,600), a hypocholesterolemia agent

A specific and sensitive radioimmunoassay (RIA) for the measurement of SQ 33,600 in biological fluids has been developed. The assay utilizes a SQ 33,600 polyclonal antibody, [125I]iodohistamine-SQ 33,600 radiolabel, and standards in serum. Satisfactory zero binding and sensitivity were obtained after an overnight incubation at 4 degrees C. Separation of the antibody-bound and free radiolabel was achieved by employing polyethylene glycol-goat anti-rabbit gamma-globulin (PEG-GARG) separant. A quantitative recovery in serum and urine of the exogenous analyte was obtained at all concentrations of SQ 33,600 tested. Intra-assay coefficients of variation (CVs) were 6.19 and 5.57% for the low and high controls, respectively. Interassay CVs were 6.64 and 6.06% for the low and medium controls, respectively. Results from the parallelism studies were acceptable for both serum and urine samples. Comparison of results from samples which were assayed by RIA and high-performance liquid chromatography (HPLC) demonstrated a significant correlation (r = 0.994; HPLC = 1.09 RIA + 57.98; n = 45). The present RIA has been successfully used to assay clinical specimens from pharmacokinetic studies.

3-Hydroxy-3-methylglutaryl-coenzyme A reductase inhibitors: oxime ether analogs of pravastatin

Pravastatin, a potent anti-hypercholesteremic drug, has been developed by Bristol-Myers Squibb for treatment of hypercholesterolemia and other related diseases. Several structurally related compounds (SQ 31,554, SQ 31,879, SQ 31,947, SQ 32,391, SQ 32,770, SQ 32,390 and SQ 32,469) modified at the 3-position of the hexahydronaphthalene ring system of pravastatin were prepared in the course of developing the basic reagents for a radioimmunoassay of the parent drug. The biological activity of these analogs was comparable to pravastatin itself. Indeed, one member of this series was found to be several times more potent than pravastatin.

High-performance liquid chromatographic-ultraviolet assay for the simultaneous quantitation of BMS-181101 and its putative hydroxy metabolites in rat and monkey plasma

A specific, accurate, precise, and reproducible High-performance liquid chromatographic-Ultraviolet (HPLC-UV) method was developed for the simultaneous quantitation of BMS-181101 (I), a new antidepressant, and its putative metabolites, 6'-hydroxy (II) and 7'-hydroxy (III) of BMS-181101 in rat and monkey plasma. The assay procedure involved solid-phase extraction of the three analytes and the internal standard (IS; BMY-42568) on 1 mL Bond Elut CN cartridge using an automated solid phase extraction controller (ASPEC) system. The final elution of the analytes was performed using 0.25% triethylamine in methanol. The eluate mixture was evaporated to dryness, the residue was reconstituted in the mobile phase and injected onto a Zorbax Phenyl column (4.6 x 250 mm; 5 microns) at a flow-rate of 1.2 mL/min. The mobile phase consisted of 20% acetonitrile, 10% methanol, 69% water and 1% 1.0 M ammonium phosphate and 1.0 M tetramethylammonium hydroxide mixture adjusted to pH 3 by phosphoric acid. An ultraviolet absorbance detector set at 287 nm was used to detect the analytes. The nominal retention times were 5, 8, 15, and 18 min for II, III, I, and IS, respectively. The standard curves for the three analytes were linear in the concentration range of 50-1000 ng/mL. The lower limit of quantitation was 50 ng/mL for each analyte. The analyses of quality control (QC) samples indicated that the nominal values could be predicted with an accuracy of (+/-) 10.5% for all three analytes in rat and monkey plasma. The precision values of the QC samples for all three analytes were within 12.7% RSD for rat and monkey plasma. All three analytes and the IS were stable in the autosampler for at least 38 h; freeze/thaw stability of the 3 analytes was established for three cycles. Stability of BMS-181101 was established for one month at -20 degrees C. The application of the assay to a pharmacokinetic study in monkey is described.

Radioimmunoassay for ceronapril, a new angiotensin-converting enzyme inhibitor, and its application to a pharmacokinetic study in healthy male volunteers

Ceronapril is a member of a new chemical class of angiotensin-converting enzyme inhibitors being developed by The Bristol-Myers Squibb Pharmaceutical Research Institute. A radioimmunoassay (RIA) has been developed for the measurement of ceronapril in biological fluids. The RIA has a range of 0 to 500 ng/ml and has the sensitivity to detect 1.0 ng/ml of ceronapril. Satisfactory zero binding and sensitivity were obtained after a 2-h incubation at room temperature or overnight at 4 degrees C. Separation of the antibody-bound and free radiolabel was achieved by employing polyethylene glycol-goat anti-rabbit gamma-globulin separant. A quantitative recovery of the exogenous analyte was obtained at all concentrations of ceronapril tested. Intraassay coefficients of variance (CV's) were 3.9% and 4.6% for the low and medium controls, respectively. A highly significant statistical correlation between RIA and [14C]TLRC was observed for both plasma and urine samples. Clinical samples from the ascending dosage studies have been analyzed by the ceronapril RIA. The maximum concentration and the area under the plasma concentration-time curve did not increase in a dose-proportional manner for doses above 100 mg.

A radioimmunoassay for the new antiviral agent 1-beta-D-arabinofuranosyl-E-5-(2-bromovinyl)uracil

A specific and sensitive radioimmunoassay (RIA) for the measurement of 1-beta-D-arabinofuranosyl-E-5-(2-bromovinyl)uracil (BV-araU) in biological fluids has been developed. The assay has a range of 2.5-1,000 ng/ml and 10-1,000 ng/ml for serum and urine, respectively, and has the sensitivity to detect 2.5 and 25 ng/ml of BV-araU in serum and urine, respectively. A satisfactory zero binding and sensitivity were obtained after an overnight incubation at 4 degrees C. Separation of the antibody-bound and free ligand was achieved by employing polyethylene glycol-goat anti-rabbit gamma globulin separant. A quantitative recovery of the exogenous analyte was obtained at all concentrations of BV-araU tested. The assay is specific for the parent drug and is not affected by the presence of its metabolite, BV-U (bromovinyl uracil) or serum components (nucleotides, nucleosides, or sugars). Intraassay coefficients of variation were 3.1-4.4% and 2.5-4.2% for serum and urine controls, respectively. Interassay variability was < 8.6% for all serum and urine controls. Linear regression analysis showed that the correlation between RIA and high-pressure liquid chromatography was excellent (r = 0.997). The ascending dosage studies have been analyzed by the BV-araU RIA, and results indicate that the values of area under the serum concentration-time curve increased proportionally with the administered dose of BV-araU up to 80 mg. Cumulative urinary excretion data showed that approximately 50% of unchanged BV-araU was excreted in the urine within 24 h.

Is High-Flow Nasal Oxygenation a Game Changer in Endobronchial Ultrasound-Guided Transbronchial Needle Aspiration?: A pilot study

Objectives

This study aimed to compare the high-flow nasal oxygen (HFNO) and supraglottic airway device (SAD) techniques in oncological patients undergoing endobronchial ultrasound (EBUS) and transbronchial needle aspiration (TBNA) to evaluate the efficacy of HFNO in them.

Methods

This pilot study was conducted at Sultan Qaboos Comprehensive Cancer Care and Research Centre, Muscat, Oman, from May 2022 to March 2023. Patients undergoing EBUS TBNA under moderate sedation were quasi-randomised into the HFNO and SAD groups. The episodes and duration of hypoxia and the lowest level of oxygen saturation were the primary outcomes measured.

Results

A total of 24 patients were included in the study (10 of them were in the HFNO group and 14 were in the SAD group), with an equal number of males and females. The duration of the procedure in both groups was similar (45 ± 20 and 44 ± 17 minutes in the HFNO and SAD groups, respectively). The mean lowest oxygen saturation in the HFNO group was 93.5 ± 4.5%, which was significantly higher than that of the SAD group (90 ± 3%; P <0.001). In both groups, maximum hypoxia occurred during the early phase of the procedure. However, the HFNO and SAD groups were similar in terms of the cumulative duration of hypotension (140 versus 95 seconds, respectively) and bradycardia (25 versus 40 seconds, respectively).

Conclusion

HFNO is a good alternative to SAD and could be used safely and efficiently in patients undergoing EBUS TBNA.

Lipid profile in acute stroke: A prospective study

The dramatic decline in the risk of stroke is most likely secondary to a better control of risk factors, especially hypertension. An increase in low-density lipoprotein (LDL) cholesterol or a decrease in high-density lipoprotein (HDL) cholesterol may be additional risk factors for cerebrovascular diseases. In this study, we prospectively evaluated patients with ischemic stroke (except for cardioembolic) and age-matched controls for serum cholesterol, triglycerides, LDL cholesterol, HDL cholesterol, very low-density lipoprotein cholesterol, and apoprotein A and B. Blood levels were determined at least 8 weeks after the acute stroke or transient ischemic attack. Between September 1989 and September 1990,59 patients and 60 controls were investigated. There were no differences among total cholesterol, triglycerides, LDL cholesterol, and apoproteins in the two groups. HDL cholesterol, however, was significantly higher in patients with stroke. In stroke patients, total cholesterol (p < 0.05), LDL cholesterol (p < 0.05), and apolipoprotein B (p < 0.05) were significantly higher in female patients compared to male patients. This study confirms that low HDL cholesterol may be a risk factor for stroke. Additionally, ally, we show that lipid abnormalities may appear to be more significant in women.