Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Linear Regression”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 631 records · Page 35Linked to original sources
PubMed2005

Long-term physiologic changes of intraocular pressure: a 10-year longitudinal analysis in young and middle-aged Japanese men.

OBJECTIVE: Although the intraocular pressure (IOP) is the most important factor related to the onset and progression of glaucoma, there is little evidence on long-term aging effects on IOP. This article examines the changes of IOP over 10 years in normal eyes to assess physiologic changes related to aging. DESIGN: Population-based long-term longitudinal study. STUDY POPULATION AND OBSERVATION PROCEDURE: Two thousand nine hundred eighty-seven Japanese male aircraft crew members underwent IOP measurement by Goldmann apparatus and received physical and complete ophthalmologic examinations every year for 10 years. A total of 2330 healthy persons (21-49 years; mean age, 35.9+/-6.8 [standard deviation]) who had no history of treatment for ophthalmic diseases, current illnesses, and no missing data for 10 consecutive years were analyzed. MAIN OUTCOME MEASURES: For analysis of the longitudinal index (trend), the linear regression coefficients for 11 points of measurement were determined. Ophthalmologic and physiologic factors affecting the 10-year mean values and a trend of IOP were examined by multivariable linear regression analysis. RESULTS: Intraocular pressure tended to decrease with age in all age groups. The mean linear regression coefficient (right eye/left eye) = -0.076/-0.060 (95% confidence interval [CI]: (-0.094 to -0.057)/(-0.078 to -0.041), -0.073/-0.060 [95% CI: (-0.084 to -0.062)/(-0.071 to -0.049)], and -0.060/-0.050 [95%CI: (-0.075 to -0.046)/(-0.064 to -0.036)] (mmHg/year) in the 20s, 30s, and 40s, respectively). In investigating correlations between the 10-year mean values of IOP and factors examined in this study, multivariate analysis showed a significantly inverse correlation with spherical power (partial regression coefficient [B] = -0.155/-0.144) and positive correlation with esophoria (B = 0.536/0.521), systolic blood pressure (B = 0.021/0.022), heart rate (B = 0.024/0.024), and hematocrit (B = 0.041/0.043) with IOP. The trend of IOP was significantly positively associated with the trends of systemic factors: body mass index (BMI) (B = 0.117/0.119), blood pressure (systolic) (B = 0.020/0.020), and hematocrit (B = 0.057/0.045), but not with any ophthalmologic factor. CONCLUSIONS: The long-term observation clearly demonstrated that, in normal eyes, the IOP decreased with aging. The IOP value was negatively associated with spherical power and positively with esophoria, blood pressure, heart rate, and hematocrit. The change of IOP could be affected mainly by the change of body mas index, blood pressure, and hematocrit.

Adult↗
PubMed1998

Effect of the number of samples on Bayesian and non-linear least-squares individualization: a study of cyclosporin treatment of haematological patients with multidrug resistance.

We have studied whether the prediction of drug concentrations improves as the number of samples used for individualization is increased, and whether the Bayesian method of individualization is superior to the non-linear least-squares method. Data were obtained from ten adult haematological patients with multidrug resistance who were treated with cyclosporin. The predictions of blood-cyclosporin concentrations were made using the Abbott PKS program. The number of samples used for individualization was increased from 1 to 30 for the Bayesian method and from 4 to 30 for the non-linear least-squares method. Linear regression, percentage prediction error, and absolute and relative predictive performance were used to evaluate the predictions. The results show that the Bayesian method affords greater precision than the non-linear least-squares method, but that the non-linear least-squares method is more accurate and results in less bias. Whereas for linear regression predictions improve as the number of samples is increased, other evaluations show improvement in the range from 5 to 11 samples; linear regression, percentage prediction errors and prediction bias support the opinion that the Bayesian method progressively becomes the non-linear least-squares method as the number of samples used for individualization is increased, but the accuracy and precision of prediction do not support this opinion. The study supports the statement that Bayes' law requires parameters from an infinite population, otherwise the advantage of the Bayesian method might be marginal.

Adult↗
PubMed2002

Limited and optimal sampling strategies for etoposide and etoposide catechol in children with leukemia.

Etoposide is used to treat childhood malignancies, and its plasma pharmacokinetics have been related to pharmacodynamic endpoints. Limiting the number of samples should facilitate the assessment of etoposide pharmacokinetics in children. We compared limited sampling strategies using multiple linear regression of plasma concentrations and clearance with Bayesian methods of estimating clearance using compartmental pharmacokinetic models. Optimal sampling times were estimated in the multiple linear regression method by determining the combination of two samples which maximized the correlation coefficient, and in the Bayesian estimation approach by minimizing the variance in estimates of clearance. Clearance estimates were compared to the actual clearances from Monte Carlo-simulated data and predicted clearances estimated using all available plasma concentrations in clinical data from children with acute lymphoblastic leukemia. Multiple linear regression poorly predicted clearance (mean bias 8.3%, precision 17.5%), but improved if plasma concentrations were logarithmically transformed (mean bias 1.4%, precision 12.5%). Bayesian estimation methods with optimal samples gave the best overall prediction (mean bias 2.5%, precision 6.8%) and also performed better than regression methods for abnormally high or low clearances. We conclude that Bayesian estimation with limited sampling gives the best estimates of etoposide clearance.

Analysis of Variance↗
PubMed2000

Analytical reliability of mobile-phase recycling in liquid chromatography

Although it is not officially condoned, mobile-phase (MP) recycling has become a widespread practice that is not well documented in terms of its effects on sample quantification. MP was spiked with three different concentrations of two analytes, tartaric acid and sodium citrate, to simulate MP recycling. These MPs were used to analyze eight different concentrations of these analytes in standard solutions. When analyte concentration in the MP exceeds that in the sample, a vacancy (negative) peak is observed, and when its concentration in the MP is equal to that in the sample, no peak is observed for that analyte. The slopes of the linear regression lines for standards in MPs with different concentrations of analyte did not change, although the gamma-intercept values decrease with increasing concentration of analyte in the MP. These results show that the concentration of analytes in recycled MP can be determined by comparing the absolute value of the gamma-intercept of the linear regression line with the corresponding peak area from the linear regression line for clean (solute-free) MP, provided other chromatographic parameters do not change. Suggestions are made for determining when recycled MP should be discarded.

Journal Article↗
PubMed2006

The effect of the development of an emergency transfer system on the travel time to tertiary care centres in Japan.

BACKGROUND: In Japan, the emergency medical system is categorized into three levels: primary, secondary, and tertiary, depending on the severity of the condition of the patient. Tertiary care centres accept patients who require 24-h monitoring. In this research, the average travel times (minutes) from the centroids of all municipalities in Japan to the nearest tertiary care centre were estimated, using the geographic information system. The systems affecting travel time to tertiary care centres were also examined. Regression analysis was performed to determine the factors affecting the travel time to tertiary care centres, using selected variables representing road conditions and the emergency transfer system. Linear regression analysis was performed to identify specific benchmarks that would be effective in reducing the average travel time to tertiary care centres in prefectures with travel times longer than the average 57 min. RESULTS: The mean travel time was 57 min, the range was 83 min, and the standard deviation was 20.4. As a result of multiple regression analysis, average coverage area per tertiary care centre, kilometres of highway road per square kilometre, and population were selected as variables with impact on the average travel time. Based on results from linear regression analysis, benchmarks for the emergency transfer system that would effectively reduce travel time to the mean value of 57 min were identified: 26% pavement ratio of roads (percentage of paved road to general roads), and three tertiary care centres and 108 ambulances. CONCLUSION: Regional gaps in the travel time to tertiary care centres were identified in Japan. The systems we should focus on to reducing travel time were identified. Further reduction of travel time to tertiary care centres can be effectively achieved by improving these specific systems. Linear regression analysis showed that a 26% pavement ratio and three tertiary care centres are beneficial to prefectures with an average time longer than the mean score, to achieve a reduction of travel time. Measures for reducing travel time need to be considered in policy-making to re-evaluate the current locations of tertiary care centres to provide equality of access to emergency medicine.

Catchment Area, Health↗
PubMed1999

The relationship between the visual analog pain intensity and pain relief scale changes during analgesic drug studies in chronic pain patients.

BACKGROUND: Most analgesic drug studies in humans quantify drug action based on verbal reports of pain intensity and pain relief. Although measures of pain intensity and pain relief show a good overall correlation, it is not known if they relate to each other consistently over time Such consistency is necessary if both measures are used to depict analgesic drug action versus time. This study examined in chronic pain patients if the relationship between visual analog pain intensity and pain relief scores was consistent during two analgesic drug studies. METHODS: Data from two independently performed analgesic drug studies were analyzed using linear regression. Data were split into pain intensity and pain relief scores recorded before and after patients' experience of maximum analgesia (>90% of maximum pain relief). The slopes of the linear regression line depicting pain intensity versus pain relief scores before and after maximum analgesia were statistically compared. RESULTS: The slope of the linear regression line before and after maximum analgesia was significantly different in both drug studies (nonoverlapping 95% confidence intervals), -2.16+/-0.57 versus -1.05+/-0.10 and -1.47+/-0.26 versus -1.09+/-0.07, respectively. These results are compatible with the observation that patients indicating the same pain intensity before and after maximum analgesia reported a different magnitude of pain relief. CONCLUSIONS: The relationship between visual analog pain intensity and pain relief scores changed systematically during both analgesic drug studies. The authors hypothesize that patients' interpretation of the pain relief scale had changed during the studies and therefore suggest using the pain intensity scale to quantify analgesic drug action over time.

Adult↗
PubMed2006

Different patterns of pulse spectrum between survivors and non-survivors during progressive hemorrhage in rats.

Previous work from our laboratory has demonstrated that the percentage differences of 2nd (C2) and 3rd (C3) pulse harmonics related to Kidney and Spleen were both increased toward another steady state in rats after acute hemorrhage. Therefore, it is suggested that changes in pulse spectra might represent the ability of animals to survive a model of progressive hemorrhage. In this study, the difference of the pulse spectra patterns between survivors and non-survivors after progressive hemorrhage (by loss of 5%, 10% or 20% of the estimated blood volume) in anesthetized rats is determined. Seven rats, dead within 2 hours after a loss of 20% of the estimated blood volume hemorrhage, were defined as 'non-survivors'. The other eleven rats, more than 2 hours after hemorrhage, were defined as 'survivors'. Pulse waves of arterial blood pressure before and after the hemorrhage were measured in parallel to the pulse spectrum analysis. Data among different phases were analyzed using one-way analysis of variance (ANOVA) with Duncan's test for pairwise comparisons. Differences between survivor and non-survivor groups at each phase were analyzed using Student's t-test. A mixed-effects linear regression model was applied to evaluate the relationship in harmonics, which significantly differed between the two groups. The study results showed that in rats, during progressive hemorrhage, the percentage differences of 2nd harmonic proportion increased significantly; however, the result failed to show any significant difference between survivors and non-survivors. After the third blood withdrawal process, the percentage differences of 3rd harmonic proportion increased more significantly in the survivors. In addition, the percentage differences of 1st harmonic proportion related to the Liver for the survivor group was significantly lower than that of the non-survivors. After analysis with the mixed linear regression model, C3 and C1 demonstrated a linear regression relationship, and there existed significant differences between survivors and non-survivors. These results suggest that C3 might play an important role in physiology regarding surviving capability after progressive hemorrhage.

Analysis of Variance↗
PubMed1985

On line determination of airway resistance by plethysmography and microcomputer.

On line determination of airway resistance is carried out by an Apple II connected to a flow body plethysmograph. Recognition of the panting maneuver is performed in real time. Linear drift of plethysmographic volume V is eliminated by taking the derivatives V and Vm of V and mouth flow Vm. Linear regression of V on Vm yields an estimate R of airway resistance, at a volume close to functional residual capacity FRC. The values of FRC is given by linear regression of V on Pm, the derivative of mouth pressure. No significant difference has been found between the estimates of the specific conductance sG calculated by linear regression and by spectral analysis of V and Vm. As variability of sG can be kept below 10%, the regression technique appears to be reliable for routine patient testing and pharmacological studies.

Airway Resistance↗
PubMed2002

Improved electrocardiographic detection of left ventricular hypertrophy.

We studied the echocardiographic measurements and the digital electrocardiograms (ECGs) of 3378 participants in the Cardiovascular Health Study and divided the subjects into a Learning Set of 1678 subjects and a demographically matched Test Set of 1700 subjects. We defined left ventricular hypertrophy (LVH) as values of the echocardiographic left ventricular mass index (LVMI) >116 g/m(2) in men or >104 g/m(2) in women. We performed univariate linear regression separately on the male and female subjects in the Learning Set to find all the ECG parameters that correlated significantly with LVMI. We then applied multivariate linear regression (MLR) to these parameters to identify the 4 variables for each sex that discriminated best between the subjects with and without LVH. We then applied the resultant sex-specific linear regressions to the subjects in the Test Set. We found that the diagnostic performances for LVH of the new MLR equations were superior to those of three commercial computerized ECG diagnostic algorithms in both men and women. We also found that for the new MLR and for all three commercial algorithms, the diagnostic sensitivities at matched specificities were superior in the men compared to those in the women. We conclude that the new MLR equation identifies patients with echocardiographically determined LVH more reliably than do 3 commercial computerized diagnostic algorithms and that the sensitivities of each of these modalities is greater in men than it is in women.

Echocardiography↗
PubMed1998

Monitoring of pH and PCO2 in children using the Paratrend 7 in a peripheral vein.

PURPOSE: To determine the accuracy of the Paratrend 7 monitor inserted into a peripheral vein in providing continuous measurements of PCO2 and pH in children with respiratory failure. METHODS: In a prospective, case series at a University based PICU in a tertiary referral hospital and following informed consent, the Paratrend 7 intra-arterial blood gas monitoring system was inserted through a peripheral intravenous cannula. When clinically indicated, an arterial blood gas (ABG) was obtained. The values of pH and PCO2 from the Paratrend 7 and the ABG were compared using linear regression analysis and Bland-Altman analysis. RESULTS: Seventeen ABGs were obtained in the 4 patients. PaCO2 ranged from 29 to 51 mmHg and the pH values from 7.16 to 7.50. Linear regression analysis for the: Paratrend 7 pH vs the ABG pH resulted in an r = 0.9678 and an r2 = 0.9367 (P < 0.0001). Linear regression analysis for the Paratrend 7 CO2 vs ABG CO2 resulted in an r = 0.8273 and an r2 = 0.6845 (P = 0.01). Using a Bland and Altman analysis, the bias +/- precision of the PCO2 values was 3 +/- 1.4 and 0.036 +/- 0.02 for the pH values. CONCLUSION: The pH and PCO2 values obtained from the Paratrend 7 monitor provide a clinically acceptable correlation with arterial values.

Adolescent↗
PubMed1997

Cardiac output estimated noninvasively from oxygen uptake during exercise.

Because gas-exchange measurements during cardiopulmonary exercise testing allow noninvasive measurement of oxygen uptake (Vo2), which is equal to cardiac output (CO) x arteriovenous oxygen content difference [C(a-vDo2),] CO and stroke volume could theoretically be estimated if the C(a-vDo2) increased in a predictable fashion as a function of % maximum Vo2 (Vo2max) during exercise. To investigate the behavior of C(a-vDo2) during progressively increasing ramp pattern cycle ergometry exercise, 5 healthy subjects performed 10 studies to exhaustion while arterial and mixed venous blood were sampled. Samples were analyzed for blood gases (pH, Pco2, Po2) and oxyhemoglobin and hemoglobin concentration with a CO-oximeter. The C(a-vDo2) (ml/100 ml) could be estimated with a linear regression [C(a-vDo2) = 5.72 + 0.105 x % Vo2max; r = 0.94]. The CO estimated from the C(a-vDo2) by using the above linear regression was well correlated with the CO determined by the direct Fick method (r = 0.96). The coefficient of variation of the estimated CO was small (7-9%) between the lactic acidosis threshold and peak Vo2. The behavior of C(a-vDo2), as related to peak Vo2, was similar regardless of cardiac function compared with similar measurements from studies in the literature performed in normal and congestive heart failure patients. In summary, CO and stroke volume can be estimated during progressive work rate exercise testing from measured Vo2 (in normal subjects and patients with congestive heart failure), and the resultant linear regression equation provides a good estimate of C(a-vDo2).

Adult↗
PubMed1986

Retinal visual acuity with pattern VEP normal subjects and reproducibility.

Retinal visual acuity was determined using an objective method by means of VEP. Sixty subjects with subjective visual acuity ranging from 20/25 to 20/12 were measured with a computer-controlled laser interferometer, using interference fringes modulated in counterphase as pattern-reversal stimulus. Steady-state VEPs taken at different spatial frequencies were evaluated by Fourier analysis and linear regression. Objective retinal acuity was calculated as the intersection between the linear regression line and the linear scaled spatial frequency axis. In 68% of all cases, the difference between objective and subjective visual acuity was less than +/- 5.1 cycles per degree (c/d) (+/- 15.4% referred to subjective acuity); in 86%, the difference was less than +/- 7.7 c/d (+/- 23.1%).

Evoked Potentials, Visual↗
PubMed2006

Relationships among body composition measures in community-dwelling older women.

OBJECTIVE: To examine whether simple anthropometric measures provide a good estimate of total and visceral fat in 146 community-dwelling, older white women (mean age, 74.0 +/- 4.1 years). RESEARCH METHODS AND PROCEDURES: Total body fat and visceral fat were measured using electron beam computed tomography (EBT). Anthropometric parameters (height, weight, BMI, sagittal diameter, and waist circumference) were measured using standard techniques. Total percentage body fat was assessed using DXA. Spearman correlations were used to examine the association between the measures. Linear regression, controlling for age, was used to examine the associations between the anthropometric parameters and total and visceral body fat measured by EBT. RESULTS: Correlations among body composition measures ranged from rho = 0.46 to 0.93 (p < 0.0001). EBT total fat was strongly correlated with both DXA estimates of total percentage fat (rho = 0.86) and BMI (rho = 0.89). Separate linear regression models indicated that BMI, waist circumference, sagittal diameter, and DXA total percentage fat were each independently related to EBT total fat. BMI had the strongest linear relationship, explaining 80% of the model variance (p < 0.0001). Linear regression indicated that BMI, waist circumference, and sagittal diameter were each independently related to EBT visceral fat, with BMI and sagittal diameter explaining approximately 53% of the model variance (p < 0.0001). DISCUSSION: The use of simple anthropometric measures such as BMI, sagittal diameter, and waist circumference may be an appropriate alternative for more expensive techniques when assessing total fat but should be used with caution when estimating visceral body fat.

Absorptiometry, Photon↗
PubMed2004

Comparison of simultaneous measurement of mouse systolic arterial blood pressure by radiotelemetry and tail-cuff methods.

Radiotelemetry of mouse blood pressure accurately monitors systolic pressure, diastolic pressure, heart rate, and locomotor activity but requires surgical implantation. Noninvasive measurements of indirect systolic blood pressure have long been available for larger rodents and now are being reported more frequently for mice. This study compared mouse systolic arterial blood pressure measurements using implanted radiotelemetry pressure transducer with simultaneous tail-cuff measurements in the same unanesthetized mice. The pressure range for comparison was extended by inducing experimental hypertension or by observations of circadian elevations between 3 AM and 6 AM. Both trained and untrained tail-cuff operators used both instruments. Every effort was made to follow recommended manufacturer's instructions. With the initial flow-based tail-cuff instrument, we made 671 comparisons (89 sessions) and found the slope of the linear regression to be 0.118, suggesting poor agreement. In an independent assessment, 277 comparisons (35 sessions) of radiotelemetry measurements with the pulse based tail-cuff instrument were made. The slope of the linear regression of the simultaneous measurements of systolic pressures was 0.98, suggesting agreement. Bland-Altman analysis also supported our interpretation of the linear regression. Thus although reliable systolic pressure measurements are possible with either tail-cuff or radiotelemetry techniques, in our hands some tail-cuff instruments fail to accurately detect elevated blood pressures. These data, however, do not distinguish whether this instrument-specific tail-cuff failure was due to operator or instrument inadequacies. We strongly advise investigators to obtain an independent and simultaneous validation of tail-cuff determinations of mouse blood pressure before making critical genotyping determinations.

Animals↗
PubMed2000

Characterisation of muscarinic receptor subtypes in avian smooth muscle.

The identity of the muscarinic receptor subtype in the chick ileum was investigated in functional and binding studies. Preliminary studies [Choo, L.-K., Mitchelson, F., Napier, P. 1988. J. Auton. Pharmacol. 8, 259-266] suggested apparent avian and mammalian family differences in the muscarinic receptor profile of ileal smooth muscle. In the current study, further characterisation was undertaken using a greater range of antagonists exhibiting high affinity for specific muscarinic receptor subtypes. Dissociation constants from functional and binding experiments were compared with published values for antagonists at each of the five muscarinic receptor subtypes. Linear regression and correlation analyses revealed the receptor initiating the contractile response was most likely of the muscarinic M(3) receptor subtype as the slope of the linear regression was 1.01+/-0.14 and the corresponding correlation coefficient (r) was 0.95. The mammalian muscarinic M(5) receptor subtype also showed a high correlation with the data giving a slope of 0.89+/-0.27 and r value of 0.76. These findings were in direct contrast to those from binding experiments in which the single binding site detected was of the muscarinic M(2) receptor subtype. The slope of the linear regression was 1.14+/-0.24 with an r value of 0.87. Thus, these results suggest that there exists a high proportion of the muscarinic M(2) receptor subtype within the tissue that does not contribute to the functional response.

Animals↗
PubMed1989

Nutrient balance in the preterm infant. 3. Effect of balance duration on outcome measurements.

To determine the effect of balance duration on outcome measurements 18 84-h balances were performed in a group of formula-fed preterm infants (birth weight = 1,362 +/- 125 g; gestational age = 32.3 +/- 1.7 weeks). Each 84-h balance was broken into contiguous 48- and 36-h periods. Comparisons (paired t tests, paired variance tests, and linear regression analysis) were made between the 48-h and 36-h (paired t and F tests) and between the 48-h and 84-h collections (simple linear regression). Nitrogen absorption (mean +/- SD) during 48-h collection (365 +/- 58 mg/kg/day and 87 +/- 3%) was not significantly different from that of the 36-h collection (381 +/- 66 mg/kg/day and 84 +/- 4%). Nitrogen retention during the 48-h collection (301 +/- 59 mg/kg/day, 72 +/- 6%) also was not significantly different from that of the 36-h collection (318 +/- 64 mg/kg/day, 72 +/- 6%). Moreover, nitrogen absorption and retention during the 48-h collection were significantly related (p less than 0.0001 for linear regression comparisons) to that of the 84-h collection. Similar observations were noted for calcium, phosphorus, magnesium, zinc, and copper absorption and retention. We conclude that a 48-h collection may be used in place of an 84-h collection without compromising the accuracy of the balance data in the enterally fed preterm infant.

Birth Weight↗
PubMed1989

Isovolumic relaxation period as an index of left ventricular relaxation under different afterload conditions--comparison with the time constant of left ventricular pressure decay in the dog.

In order to determine whether isovolumic relaxation period (IRP) reflects left ventricular relaxation under different afterload conditions, 17 anesthetized, open chest dogs were studied, and the left ventricular pressure decay time constant (T) was calculated. In 12 dogs, angiotensin II and nitroprusside were administered, with the heart rate constant at 90 beats/min. Multiple linear regression analysis showed that the aortic dicrotic notch pressure (AoDNP) and T were major determinants of IRP, while left ventricular end-diastolic pressure was a minor determinant. Multiple linear regression analysis, correlating T with IRP and AoDNP, did not further improve the correlation coefficient compared with that between T and IRP. We concluded that correction of the IRP by AoDNP is not necessary to predict T from additional multiple linear regression. The effects of ascending aortic constriction or angiotensin II on IRP were examined in five dogs, after pretreatment with propranolol. Aortic constriction caused a significant decrease in IRP and T, while angiotensin II produced a significant increase in IRP and T. IRP was affected by the change of afterload. However, the IRP and T values were always altered in the same direction. These results demonstrate that IRP is substituted for T and it reflects left ventricular relaxation even in different afterload conditions. We conclude that IRP is a simple parameter easily used to evaluate left ventricular relaxation in clinical situations.

Angiotensin II↗
PubMed2002

Urinary N-acetyl-beta-glucosaminidase excretion is a marker of tubular cell dysfunction and a predictor of outcome in primary glomerulonephritis.

BACKGROUND: The urinary excretion of N-acetyl-beta-glucosamynidase (NAG) is increased in subjects exposed to substances toxic for renal tubular cells. In experimental and human glomerular diseases, its increased excretion is probably due to the dysfunction of tubular epithelial cells induced by increased traffic of proteins in the tubular lumen. The first aim of this study was to evaluate whether NAG excretion is correlated not only with the amount of proteinuria but also with some proteinuric components which reflect both glomerular capillary wall damage (IgG) and an impairment of tubular reabsorption of microproteins (alpha(1) microglobulin). The second aim was to assess whether NAG excretion has a predictive value on functional outcome and response to therapy. METHODS: In 136 patients with primary glomerulonephritis [74 with idiopathic membranous nephropathy (IMN), 44 with primary focal segmental glomerulosclerosis (FSGS) and 18 with minimal change disease (MCD)] urinary NAG excretion was measured by a colorimetric method and expressed in units per gram of urinary creatinine. RESULTS: Using univariate linear regression analysis NAG excretion in all 136 patients was significantly dependent on IgG excretion, 24-h proteinuria, fractional excretion of alpha(1) microglobulin (FE alpha(1)m) and diagnosis. Using multiple linear regression analysis, NAG excretion was significantly dependent only on IgG excretion and 24-h proteinuria. Limiting the analysis to 67 patients with nephrotic syndrome (NS) and baseline normal renal function, by multiple linear regression, NAG excretion was significantly dependent on IgG excretion (P=0.0004), 24-h proteinuria (P=0.0067) and FE alpha(1)-m (P=0.0032) (R(2)=0.63). In 66 patients with NS and normal baseline renal function (MCD 10 patients; FSGS 20 patients; IMN 36 patients), according to values below or above defined cut-offs (IMN, </= or >18 U/g urinary Cr; FSGS and MCD, </= or >24 U/g urinary Cr), NAG excretion predicted remission in 86 vs 27% of IMN patients (P=0.0002) and 77 vs 14% of FSGS patients (P=0.005). Progression to chronic renal failure (CRF) was 0 vs 47% in IMN patients (P=0.001) and 8 vs 57% in FSGS patients (P=0.03). Using Cox model, in IMN patients only NAG excretion (P=0.01, RR 5.8), but not 24-h proteinuria, predicted progression to CRF. All MCD patients had NAG excretion values below the chosen cut-off, and 90% of them developed remission. Response to immunosuppressive therapy was significantly different in patients with NAG excretion values below or above the cut-offs. CONCLUSION: Urinary NAG excretion can be considered as a reliable marker of the tubulo-toxicity of proteinuria in the early stage of IMN, FSGS and MCD; the excretion values show a significant relationship with 24-h proteinuria, IgG excretion and FE alpha(1)m. Its determination may be a non-invasive, useful test for the early identification of patients who will subsequently develop CRF or clinical remission and responsiveness to therapy.

Acetylglucosaminidase↗