SEARCH · PubMed Health

Results for “multivariate analysis”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

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

At least 649 records · Page 36Linked to original sources

PubMed2005

Prevention of bleeding after islet transplantation: lessons learned from a multivariate analysis of 132 cases at a single institution.

Islet transplantation is being offered increasingly for selected patients with unstable type 1 diabetes. Percutaneous transhepatic portal access avoids a need for surgery, but is associated with potential risk of bleeding. Between 1999 and 2005, we performed 132 percutaneous transhepatic islet transplants in 67 patients. We encountered bleeding in 18/132 cases (13.6%). In univariate analysis, the risk of bleeding in the absence of effective track ablation was associated with an increasing number of procedures (2nd and 3rd procedures with an odds ratio (OR) of 9.5 and 20.9, respectively), platelets count <150,000 (OR 4.4), elevated portal pressure (OR 1.1 per mm Hg rise), heparin dose > or =45 U/kg (OR 9.8) and pre-transplant aspirin (81 mg per day) (OR 2.6, p = 0.05). A multivariate analysis further confirmed the cumulative transplant procedure number (p < 0.001) and heparin dose > or =45 U/kg (p = 0.02) as independent risk factors for bleeding. Effective mechanical sealing of the intrahepatic portal catheter tract with thrombostatic coils and tissue fibrin glue completely prevented bleeding in all subsequent procedures (n = 26, p = 0.02). We conclude that bleeding after percutaneous islet implantation is an avoidable complication provided the intraparenchymal liver tract is sealed effectively.

Acute Disease↗

PubMed2004

Head and neck soft tissue sarcomas: a multivariate analysis of outcomes.

BACKGROUND: Soft tissue sarcomas of the head and neck region represent a rare group of tumors of which a limited number of published individual- and institution-based experiences exist. METHODS: We performed an analysis of head and neck sarcoma patients identified from our institution between 1973 and 1999. Exclusion criteria included pediatric rhabdomyosarcomas, sarcomas of the neuromeningeal axis or non-head and neck primary disease sites, and bone sarcomas. All cases underwent pathologic re-review before statistical analysis. RESULTS: After pathologic review, 111 head and neck sarcoma patients remained (mean age, 47 +/- 20 years). The median duration of follow-up was 51 months; the actuarial 5-year relapse-free, disease-specific, and overall survivals were 55%, 52%, and 44%, respectively. Forty-six percent remained free of recurrence at the most recent follow-up, and the most common site of recurrence was local followed by distant sites. By multivariate analysis, size and grade significantly influenced relapse-free, disease-specific, and overall survivals, whereas margin status additionally influenced relapse-free survival. Subset analysis of the fibrosarcoma/malignant fibrous histiocytoma and desmoid/dermatofibrosarcoma protuberans histologies was undertaken. CONCLUSIONS: Size >5 cm and high-grade histology are considered poor prognostic indicators. Patients with either of these characteristics should be considered for adjuvant trials.

Adult↗

PubMed1990

[Study on prognosis of threatened abortion assessed by multivariate analysis].

The serum concentrations of various hormones which included hPL, progesterone(P), estradiol(E2) and 17 alpha-hydroxyprogesterone (17P) in serum, and hCG in urine were measured simultaneously in patients with threatened abortion to evaluate the prognosis. Ultrasound scanning was also undertaken. One hundred and thirty-six patients at 5 to 8 weeks of gestation were divided into two groups retrospectively. Group A consisted of 99 patients with a good prognosis and group B consisted of 37 patients who aborted spontaneously. The mean concentrations of hormones, and ultrasound findings (GS, CRL and FHM) were not significantly different in both groups at 5 weeks. An attempt was made to apply the data to the computation in multivariate analysis. By discriminant analysis, 100% reliability was achieved in separating the two groups according to hormone levels only at this stage of pregnancy. At 6 weeks, the hCG, P and E2 values were statistically higher in group A than B. Because of the overlapping of the data, principal component analysis was attempted. Two groups were roughly separated. The values for factor loading indicated that GS, CRL and 17P effective in making the separation possible. At 7 and 8 weeks of gestation, the mean concentrations of hCG, hPL, P and E2 were significantly higher in A than B. The factor loading values indicated that ultrasound findings were essential in separating into two groups. These results suggested that hormonal analysis may be useful in assessing the prognosis in threatened abortion in the early stage of gestation (5 and 6 weeks), while ultrasound findings make an accurate prediction for the prognosis possible after 7 weeks of gestation.

Abortion, Threatened↗

PubMedDate not supplied

Impact on outcome of additional microvascular anastomosis--supercharge--on colon interposition for esophageal replacement: comparative and multivariate analysis.

The impact on the outcome of an additional microvascular anastomosis--supercharge--on colon interposition for esophageal replacement was retrospectively evaluated by comparing it with colon interposition without supercharge. A series of 53 patients had undergone colon interposition for esophageal replacement at Kurume University Hospital from 1981 to 1996. The postoperative courses and the morbidity and mortality rates were compared between the 24 patients who underwent colon interposition without supercharge from 1981 to 1988 and the other 29 patients who underwent colon interposition with supercharge from 1989 to 1996. Risk factors for leakage of the esophagocolostomy and for hospital mortality after colon interposition were evaluated by multivariate analysis. Colon interposition with supercharge required a longer operation time but resulted in a lower incidence of necrosis in the colon graft and leakage in the esophagocolostomy (Odds ratio = 34), a shorter duration until peroral intake, and a shorter hospital stay compared to colonic interposition without supercharge. The addition of supercharge to colon interposition for esophageal replacement has been an effective option that has prevented serious complications caused by graft ischemia.

Aged↗

PubMed1990

Multivariable analysis of serum 3,5,3'-L-triiodothyronine concentration in patients of diabetes mellitus by blood glucose level and body weight.

Serum thyroid hormone concentrations, 1-thyroxine (T4), free T4 and 3,5,3'-l-triiodothyronine (T3) were measured in 213 patients of diabetes and analyzed their correlation with metabolic parameters, hyperglycemia and body weight. Haemoglobin A1 (HbA1) was used as an index of hyperglycemia. Body weight was expressed by relative body weight (body weight/standard weight). Among the thyroid hormones, only T3 had significant correlation with HbA1 and body weight (r = -0.476, P less than 0.01 and r = 0.369, P less than 0.01, respectively). Multivariable analysis of serum T3 by HbA1 and relative body weight gave the following regression equation. Serum T3 (ng/dl) = 108 + 0.362 x relative body weight (%) - 3.88 x HbA 1 (%). Though relative body weight had inverse correlation with HbA1, the contribution of the two metabolic parameters to the serum T3 was independent from each other. Our results confirm the previous reports that low T3 in diabetes correlates with severity of hyperglycemia and we report for the first time that serum T3 of diabetic patients has positive correlation with body weight, probably due to still available carbohydrate in spite of disturbances in the metabolism.

Adult↗

PubMed1995

Mortality in sleep apnea patients: a multivariate analysis of risk factors.

During 1976-1988 we diagnosed sleep apnea syndrome (SAS) in 1,620 adult men and women monitored in the Technion sleep laboratories. Their age at the time of diagnosis ranged between 21 and 79 years. Fifty-seven patients (53 men and 4 women) had died by 1990, 53% due to respiratory-cardiovascular causes. The observed/expected (O/E) mortality rates, calculated for men only, revealed excess mortality of patients under 70 years old. Excess mortality was significant in the fourth and fifth decades (3.33, p < 0.002; 3.23, p < 0.0002, respectively). In patients older than 70 O/E was 0.33 (p < 0.0007). Hierarchical multivariate analysis with four fixed variables [age, body mass index (BMI), hypertension and apnea index] and four additional variables added manually one at a time (heart disease, lung disease, diabetes, apnea duration) was used to determine the predictors of death from all causes, cardiopulmonary causes and from myocardial infarction (MI). All four major variables were found to be significant predictors of mortality from all causes, in addition to lung disease and heart disease. Only age and BMI were significant predictors of cardiopulmonary deaths in addition to lung disease. Age, BMI and hypertension predicted MI deaths in addition to lung disease. These results were interpreted to suggest that SAS affects death indirectly, most probably by being a risk factor for hypertension.

Adult↗

PubMed2001

Predictive factors of Barrett esophagus: multivariate analysis of 502 patients with gastroesophageal reflux disease.

HYPOTHESIS: Risk factors for the presence and extent of Barrett esophagus (BE) can be identified in patients with gastroesophageal reflux disease (GERD). DESIGN: Case-comparison study. SETTING: University tertiary referral center. PATIENTS: Five hundred two consecutive patients with GERD documented by 24-hour esophageal pH monitoring and with complete demographic, endoscopic, and physiological evaluation, divided in groups according to the presence and extent of BE (328 patients without BE and 174 with BE [67 short-segment BE and 107 long-segment BE]). MAIN OUTCOME MEASURES: Clinical, endoscopic, and physiological data, studied by multivariate analysis, to identify the independent predictors of the presence and extent of BE. RESULTS: Seven factors were identified as predictors of BE. They were abnormal bile reflux (odds ratio [OR], 4.2; 95% confidence interval [CI], 1.9-9.7), hiatal hernia larger than 4 cm (OR, 4.1; 95% CI, 2.1-8.0), a defective lower esophageal sphincter (OR, 2.7; 95% CI, 1.4-5.4), male sex (OR, 2.6; 95% CI, 1.6-4.3), defective distal esophageal contraction (OR, 2.2; 95% CI, 1.4-3.5), abnormal number of reflux episodes lasting longer than 5 minutes (OR, 2.2; 95% CI, 1.1-4.6), and GERD symptoms lasting for more than 5 years (OR, 2.1; 95% CI, 1.4-3.2). Only abnormal bile reflux (OR, 4.8; 95% CI, 1.7-13.2) was identified as a predictor of short-segment BE (baseline, no BE). Three factors were identified as predictors of long-segment BE (baseline short-segment BE). They were hiatal hernia larger than 4 cm (OR, 17.8; 95% CI, 4.1-76.6), a defective lower esophageal sphincter (OR, 16.9; 95% CI, 1.6-181.4), and an abnormal longest reflux episode (OR, 8.1; 95% CI, 2.8-24.0). CONCLUSIONS: Among patients with GERD, specific factors are associated with the presence and extent of BE. Elimination of reflux with an antireflux operation in patients with 1 or more of these factors may prevent the future development of BE.

Adolescent↗

PubMedDate not supplied

Carcinoma of the tonsillar region: a multivariate analysis of 243 patients treated with radical radiotherapy.

Between October 1964 and June 1990, 243 patients with squamous cell carcinoma of the tonsillar region were treated with continuous-course radical radiotherapy at the University of Florida. All patients had a minimum 2-year follow-up. Patients were staged according to the AJCC system; stage IV was stratified into two subsets: favorable, IVA (T1-T3 or N2A-N3A); and unfavorable, IVB (T4 or N3B). The initial and ultimate local control rates (including surgical salvage) according to T stage were as follows: T1, 87% and 100%, T2, 79% and 92%; T3, 71% and 76%; and T4, 44% and 48%. Multivariate analysis demonstrated that anterior extension of the primary tumor (p = .0001) and T stage (p = .014) were the most significant parameters affecting local control after radiotherapy. For T4 lesions, twice-daily irradiation significantly improved local control (p = .04). The 5-year absolute and cause-specific survival rates as a function of modified AJCC stage were as follows: I, 37% and 100%; II, 55% and 90%; III, 55% and 85%; IVA, 35% and 60%; and IVB, 23% and 38%. The probability of a severe complication was 3% for the entire group of patients.

Brachytherapy↗

PubMedDate not supplied

Predictive value of preoperative carcinoembryonic antigen levels for the prognosis of patients with well-differentiated gastric cancer. A multivariate analysis.

Serum carcinoembryonic antigen (CEA) levels were determined preoperatively in 221 patients with well-differentiated gastric cancer. The mean preoperative serum CEA level was 15.9 +/- 88.5 ng/ml (1.0-1,133.0 ng/ml) for all patients, and the incidence of an elevated CEA (> 5 ng/ml) was 11.8% (26/221). The CEA-positive patients had larger tumors, a more prominent serosal invasion, more frequent lymphatic and vascular involvement, less expansive tumor growth and higher rates of lymph node and hepatic metastases than did the CEA-negative patients. Thus, the CEA-positive patients had a more advanced stage of disease, and 61.5% underwent noncurative resection (vs. 11.3% in CEA-negative patients). The survival rate of the CEA-positive patients was lower than that of the CEA-negative ones (p < 0.01). As the multivariate analysis revealed the preoperative CEA level to be an independent prognostic factor for survival, an assay for this antigen prior to surgery is to be recommended.

Adenocarcinoma↗

PubMed2006

Imaging markers of mild cognitive impairment: multivariate analysis of CBF SPECT.

This study aimed to investigate cross-sectional and longitudinal changes of regional cerebral blood flow (rCBF) in preclinical dementia using single photon emission computed tomography (SPECT). SPECT and cognitive function were investigated in 39 mild cognitive impairment (MCI) subjects and 20 age-matched controls. All subjects were followed longitudinally 19 months on average, 16 MCI subjects progressed to Alzheimer's disease (AD), who were retrospectively defined as progressive mild cognitive impairment (PMCI) at baseline and 23 MCI subjects remained stable and were defined as stable mild cognitive impairment (SMCI) at baseline. SPECT was performed both at the initial investigation and at follow-up. Image data were analyzed using multivariate analysis, SPM and volume of interest (VOI)-based analysis. Significant covariate patterns were derived, which differentiate among PMCI, SMCI and controls at baseline as well as describe the longitudinal progression of PMCI. The combined SPECT and neuropsychology increased the diagnostic accuracy of PMCI at baseline. SPECT and neuropsychological testing can be used objectively for both baseline diagnosis and to monitor changes in brain function during very early AD.

Aged↗

PubMed1993

Prospective randomized study of routine intraoperative cholangiography during open cholecystectomy: long-term follow-up and multivariate analysis of predictors of choledocholithiasis.

A prospective randomized study was performed to assess the value of routine intraoperative cholangiography (IOC) during cholecystectomy for gallstone disease. Four hundred and fifty-seven consecutive patients were screened for the presence of 11 predefined clinical criteria assumed to indicate choledocholithiasis. Two hundred and eighty patients who had no positive criteria and in whom preoperative endoscopic retrograde cholangiography had not been performed were randomized at the operating table to the IOC or no-IOC group. Follow-up was performed 6 to 8 years after the operation with a questionnaire and by use of clinical, biochemical, and radiologic investigations as indicated. Multivariate analysis was used to identify independent predictors of choledocholithiasis and the combination of criteria having the best predictive ability. The frequency of common bile duct calculi at operation was significantly correlated with age and with all clinical criteria except recent or present pancreatitis. However, only serum bilirubin level, cystic duct diameter, demonstration of common bile duct calculi on preoperative imaging or intraoperative palpation, and age at operation were independent predictors of choledocholithiasis. The overall best subset of clinical indicators contained all criteria with the exception of pancreatitis and alkaline phosphatase level. Negative predictive ability of the set of criteria was 100% for patients up to 60 years of age and 97% for patients older than 60 years at the time of operation. No case of residual common bile duct calculi was present in the IOC and no-IOC groups at follow-up. Our data strongly support a policy of performing IOC during cholecystectomy only when clinical criteria suggest the presence of common bile duct abnormalities or to clarify ductal anatomy.

Biliary Tract↗

PubMed1998

Multivariate analysis of survival after malfunctioning biological and mechanical prosthesis replacement.

BACKGROUND: Results of valvular reoperations depend on extrinsic and patients' intrinsic risk factors. New prosthetic substitutes continue to appear and the clinical effect is difficult to evaluate. Randomized studies are limited by patient selection and follow-up time. We followed the patient-centric outcome research applied to a large database of valvular operations. METHODS: Between January 1, 1970 and January 1, 1995 755 patients underwent one reoperation, 96 a second reoperation, and 12 a third reoperation. On January 1, 1996 a common closing date follow-up was obtained in 98.7% of reoperated patients. Multivariable analysis in the hazard domain was applied to obtain an upgradable model of survival that could be used for predictions and treatment comparison. RESULTS: Postoperative death hazard showed an early phase merging within 6 months with a constant low hazard phase. The survival proportion was 0.65 (70% CL, 0.63 to 0.67) at 5 years, 0.51 (70% CL, 0.49 to 0.53) at 10 years, 0.47 (range, 0.44 to 0.49) at 15 years, 0.42 (70% CL, 0.39 to 0.46) at 20 and 25 years. Significant incremental risk factors for early mortality were reoperative era 1970 to 1980 (hazard ratio = 2.8), reoperation number (hazard ratio = 1.9), heart penetration on surgery (hazard ratio = 7.6), emergent operation (hazard ratio = 5.8), urgent operation (hazard ratio = 2.1), prosthetic thrombosis (hazard ratio = 2.4), acute prosthetic endocarditis (hazard ratio = 3.0), acute endocarditis of the natural valve at antecedent operation (hazard ratio = 3.2), original floppy valve pathology (hazard ratio = 3.2), and mitroaortic replacement (hazard ratio = 5.7). Isolated mitral reoperation had a lower risk (hazard ratio = 0.5). Significant incremental risk factors for constant phase were: operative era (1970 to 1980) (hazard ratio = 2.0), congestive heart failure (hazard ratio = 2.6), reoperation on tricuspid valve after previous mitral insertion (hazard ratio = 4.9), reoperation for recurring dehiscence (hazard ratio = 4.6), double-valve procedure (hazard ratio = 1.6), coronary artery bypass graft (hazard ratio = 2.7), aortic root disease at original operation (hazard ratio = 2.1), older operative age (hazard ratio = 1.1). Use of bileaflet prosthesis was found to decrease significantly (p = 0.0002) the death risk (hazard ratio = 0.2). CONCLUSIONS: There is no late uprising hazard, and surviving patients remain exposed to a low risk of death (4% of patients per year). Considering simultaneously the confounding from operative age and operative era and the many concomitant risk factors, survival appears favorably influenced by use of bileaflet valves on reoperation.

Adolescent↗

PubMed2002

Histopathologic effects of neoadjuvant therapies for advanced squamous cell carcinoma of the esophagus: multivariate analysis of predictive factors and p53 overexpression.

In 97 patients (60, chemotherapy; 22, chemoradiotherapy; 15, radiotherapy), histopathologic effects were evaluated microscopically, and histologic response rates were compared among three neoadjuvant treatment modalities. Predictive factors for neoadjuvant therapies were analyzed by logistic regression, including the results of p53 immunohistochemical staining. In the chemoradiotherapy group, the pathologic response rate was 86.4%, and was significantly higher than that for chemotherapy (P < 0.0001) or for radiotherapy (P = 0.0031). In patients with normal p53 protein expression, the histopathologic response rate to chemotherapy was 20.0%, a higher rate than that for patients with abnormal p53 overexpression. In the chemoradiotherapy or radiotherapy group, however, the response rates were almost the same, irrespective of p53 oncoprotein status. From multivariate analysis, the neoadjuvant treatment modality itself was identified as the most powerful predictive factor for the effect. Chemoradiotherapy had the most powerful effect on advanced esophageal cancer, and p53 status did not influence the clinical outcome in this group.

Adult↗

PubMed1993

Uni- and multivariate analysis of eight indications for post-operative radiotherapy and their significance for local-regional cure in advanced head and neck cancer.

Eighty consecutive patients with advanced head and neck cancer were treated with combined therapy of radical surgery and post-operative radiotherapy. Indications for post-operative radiotherapy were pathological staging of the primary tumour (pT3 or pT4), poorly differentiated or undifferentiated carcinoma (G3, G4), tumour thickness, tumour invasion in the surrounding tissues with slender tumour strands and solitary tumour cells, perineural spread, lymphangio-invasive tumour growth, multiple positive neck nodes or extranodal spread, and microscopical irradicality of the surgical margins. The contribution for prognosis of these indications for post-operative radiotherapy were retrospectively calculated in an univariate and multivariate analysis. Of all investigated parameters, the mode of tumour invasion and lymphangio-invasive growth were independent prognostic factors. If these unfavourable prognostic signs are present, post-operative radiotherapy has to be intensified to at least curative doses of 66 Gy or more to the areas at risk.

Combined Modality Therapy↗

PubMed1996

[Resection of liver metastases of colorectal tumors. A uni- and multivariate analysis of prognostic factors].

Between September 1985 and December 1994, liver resections with curative intent were performed for liver metastases of colorectal primaries in 76 patients at the department for abdominal surgery of the university of Mainz. Perioperative morbidity and 30-day mortality rates were 30.3% and 6.6% respectively. The 3- and 5-year survival rate was 53% and 36% respectively. For patients with complete tumor removal (R0 resection), excluding perioperative mortality, the following factors were associated with more prolonged survival after hepatic resection in univariate analysis. N1 stage of the colorectal primary compared to N2 stage (in patients with positive mesenterial lymph nodes); diameter of the largest metastasis between 2.1 and 3.5 cm; postoperative normalized CEA level (in patients with CEA level elevated to more than 4 ng/dl preoperatively); and wedge resections compared to hemihepatectomies. Regarding disease-free survival, only N1 stage of the colorectal primary compared to N2 stage and diameter of the largest metastasis between 2.1 and 3.5 cm had a positive influence. In multivariate analysis, the diameter of the largest metastasis was the only factor associated with survival time, while disease-free survival was influenced by the largest diameter of the metastasis, patient sex and N stage of the colorectal primary. Whether adjuvant therapies (systemic or regional chemotherapy) after curative resection of colorectal liver metastases is associated with better survival times remains an open question. Some of the above-mentioned prognostic factors may be helpful in selecting patients for entry into adjuvant therapy protocols.

Adult↗

PubMedDate not supplied

Risk factors for late recurrent rejection after heart transplantation: a multiinstitutional, multivariable analysis. Cardiac Transplant Research Database Group.

BACKGROUND: Previous studies of allograft rejection have focused on early episodes and risk factors from pretransplant variables. METHODS: This multiinstitutional study compared early (< 1 year) and late (> 1 year) rejection episodes and risk factors for recurrent rejection from variables both before and after transplantation among 1251 patients who underwent primary heart transplantation and available follow-up of greater than 1 year. RESULTS: There were a total of 1882 rejection episodes over a mean follow-up of 26 +/- 0.3 months. The hazard function (instantaneous risk per patient per month) peaked at 1 month followed by a low constant risk of rejection after 12 months. By multivariable analysis, the most dominant risk factors for recurrent rejection during the first posttransplantation year were a shorter time interval since transplantation and a shorter time since a previous rejection episode. Other factors included young age, female gender, female donor, positive cytomegalovirus serology, prior infections, and OKT3 induction. In contrast, after the first year, the dominant risk factors for rejection were a greater number of rejections during the first year and the presence of prior cytomegalovirus infections. CONCLUSIONS: These data show a striking time dependency for rejection episodes among heart transplant recipients. Factors that increase risk for rejection in the first year differ from risk factors for rejection in subsequent years. These data suggest that it may be possible to tailor rejection surveillance protocols and immunosuppression intensity, according to specific patient and time-related risk factors.

Adolescent↗

PubMed1994

A multivariate analysis of assessment of myometrial invasion of endometrial carcinoma by magnetic resonance imaging.

Thirty-nine cases of endometrial carcinoma were analyzed by magnetic resonance imaging (0.15 or 1.5 tesla (T) MRI) prior to surgery to evaluate the depth of myometrial invasion of endometrial carcinoma. Several diagnostic parameters of MRI were set: maximum area of high-intensity area (HIA) in sagittal (HIASI) and transverse (HIATI) images of the uterus, minimum width of remaining myometrium (MWRM), and ratio of minimum to maximum width of normal myometrial thickness (RMMT). In analysis of correlation between myometrial invasion and MRI parameters, the correlation coefficient values of HIASI, HIATI, and MWRM were higher than those of other parameters at 0.71, 0.69, and -0.74, respectively. When cases were classified into three groups according to tumor invasion (tumor limited to endometrium, invasion to less than 50% of the myometrium, invasion to greater than 50% of the myometrium), MWRM was the most useful as a single parameter. Since HIASI and MWRM were independent parameters of myometrial invasion, they were selected for multivariate analysis. An equation to calculate the predicted depth of myometrial invasion was derived using both HIASI and MWRM by the multiple linear regression model. The equation could estimate the degree of myometrial invasion from the parameters of 1.5 T MRI. The values obtained from the equation are more useful for prediction of myometrial invasion of endometrial carcinoma than single parameters in both 0.15- and 1.5-T MRI.

Adult↗

PubMed2006

Factors associated with pulmonary edema in severe scorpion sting patients--a multivariate analysis of 428 cases.

OBJECTIVE: To determine clinical and routine laboratory factors associated with pulmonary edema secondary to scorpion envenomation. DESIGN AND SETTING: Retrospective study covering 13 years (1990-2002) in the medical Intensive Care Unit of a university hospital (Sfax-Tunisia). PATIENTS: 428 patients older than 3 years who were admitted to the intensive care unit for scorpion envenomation were included in this study. Patients were stratified into two groups according to the presence or absence of pulmonary edema as assessed by a medical committee. MEASUREMENTS AND RESULTS: The mean (+/- SD) age was 17.5 +/- 17.7 years, ranging from 3 to 76 years. The pulmonary edema group included 294 patients (68.7%). A multivariate analysis found the following factors to be correlated with a pulmonary edema: age less than 5 years (p = 0.04), sweating (p = 0.004), agitation (p = 0.01), leukocytes of 25000 cells/mm3 or more (p = 0.02), and a plasma protein concentration of 72 g/L or more (p < 0.0001). In addition, a plasma protein concentration of 72 g/L or more predicted the presence of pulmonary edema with a sensitivity of 78% a specificity of 88%, a positive predictive value of 93%, and negative predictive value of 64%. Almost 84% of patients having a respiratory rate of > or = 30 breaths/minute associated with agitation and sweating were classified in the pulmonary edema group. This clinical association indicates the presence of pulmonary edema with a specificity of 84.3% and a positive predictive value of 87.5%. CONCLUSION: In scorpion envenomation patients older than 3 years, the association of a respiratory rate of > or = 30 breaths/minute, agitation, sweating, or the presence of high plasma protein concentrations suggest the presence of pulmonary edema.

Adolescent↗