Prospective study on diagnostic and prognostic significance of postoperative FDG PET/CT in recurrent colorectal carcinoma patients: comparison with MRI and tumor markers.

Current guidelines for follow-up after resection of colorectal cancer (CRC) recommend regular measurements of carcinoembryogenic antigen (CEA) and imaging tests. Multidetector computed tomography (MDCT) and magnetic resonance imaging (MRI) are currently primary imaging modalities, while the role of fluorine-18-fluoro-deoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT), which is recommended in patients with negative MDCT and increased CEA, is still uncertain. Our aim was to compare diagnostic performance and prognostic significance of 18F-FDG PET/CT with MRI and tumor markers CEA and carbohydrate antigen 19-9 (CA 19-9) in detection of recurrent CRC. This prospective study included 35 patients with resected CRC, referred to 18F-FDG PET/CT examination for suspected recurrence. During median follow-up of 24.4±1.5 months 18F-FDG PET/CT and MRI results and tumor marker levels were compared with findings of histopathological examination or with results of clinical and imaging follow-up. Management plan before the 18F-FDG PET/CT scan was considered and compared to the final treatment decision. The sensitivity, specificity, positive and negative predictive value and accuracy of 18F-FDG PET/CT and MRI in detection of recurrent colorectal cancer in patient-based analysis were 92.6%, 75%, 92.6%, 75% and 88.6%, and 65.4%, 66.7%, 85%, 40% and 65.7%, respectively. In lesion-based analysis the sensitivity of 18F-FDG PET/CT and MRI was 83.1% and 68.2%, respectively. The overall accuracy of CEA and CA 19-9 in recurrence detection was 48.6% and 54.3%, respectively. PET/CT induced therapy alterations in 13/35 (37.1%) patients. Progression was observed in 16/35 patients during follow-up, with significantly lower risk of progression in patients with treatment changes based on PET findings (Multivariate Cox regression; p=0.017). In addition, elevated CA 19-9 levels in time of PET scan and male gender carried significantly higher risk of progression (p=0.007 and p=0.016, respectively). Kaplan-Meier Log rank test showed significantly longer progression-free survival time in patients who had treatment plan changed based on PET/CT (p=0.046). We can conclude that 18F-FDG PET/CT showed better sensitivity and accuracy compared to MRI in detection of recurrent colorectal cancer, with much better sensitivity compared to CEA and CA 19-9. Patients with treatment changes based on 18F-FDG PET/CT had significantly better prognosis and longer progression-free survival, while elevated values of CA 19-9 and male gender were associated with worse prognosis.

Metabolic activity of breast cancer metastatic lesions on positron emission tomography/computed tomography: comparison with histological and biological characteristics of primary tumor.

Higher intensity of FDG uptake on PET/CT in primary tumor is seen in patients with IDC compared to ILC, also in high grade tumours, tumours with negative ER and higher Ki67 values, while data are inconsistent in case of relation between primary tumor's PgR and HER2 expression with its metabolic activity levels. On account of the lack of studies that include research of breast cancer metastatic lesion metabolism level and its relation to tumor histology and biology, our goal was to investigate the association of metastatic lesions' glucose metabolism level on PET/CT with different histological and biological characteristics of primary tumor. In a total number of N=100 patients, highest SUVmax values for each patient were used in testing difference between metastatic metabolic activity in patients with different tumor histology, grade, ER, PgR and HER2 status, subtype, as well in testing relation of Ki67 index to metastasis' metabolism level. In testing difference between histological types of breast cancer, SUVmax values were also compared separately for each specific anatomical site (regional and distant lymph nodes, bones and liver). No difference was found regarding metastatic SUVmax values in patients with primary IDC (n=55, median SUVmax 9.70) and ILC (n=34, median SUVmax 7.20) independently of anatomic site, and for each of analysed sites separately. No difference was found as well between SUVmax detected in metastasis in patients with different grade (grade II: n=58, median SUVmax 7.70; grade III: n=12, median SUVmax 10.20), ER (59 positive, median SUVmax 8.50; 22 negative, median SUVmax 8.05), PgR (55 positive, median SUVmax 8.50; 23 negative, median SUVmax 7.80), and HER2 (14 positive, median SUVmax 6.84; 51 negative, median SUVmax 8.63) expression in primary tumor, and between patients with different tumor subtype. Ki67 was also not associated with tumor metastatic SUVmax values (n=11, rs = -0.21, p=0.53). We conclude that there is no association of primary breast cancer histological type, grade, ER, PgR, HER2 and Ki67 expression with metabolic activity in metastasis detected on PET/CT.

Comparison of 18F-FDG PET/CT and MDCT for staging/restaging of non-small cell lung cancer.

UNLABELLED: Multi-detector computed tomography (MDCT) is most commonly used for staging of non-small cell lung cancer (NSCLC). In recent years, 18F- fluorodeoxyglucose positron emission tomography combined with computed tomography (18F-FDG PET/CT) has also been used for the same purpose. Since studies comparing these two methods are scarce, our aim was to determine how the TNM classification and thereby staging of NSCLC compare between 18F-FDG PET/CT and MDCT. 18F-FDG PET/CT and MDCT were collected in 83 patients with NSCLC 3 to 30 days apart (median 17 days). The investigators interpreting 18F-FDG PET/CT were unaware of MDCT results. The Cohen's kappa (κ) was calculated to determine the rate of agreement. The hypothesis was that the strength of agreement between the two methods will be at least moderate (κ>0.40) based on the adopted criteria (κ<0.20 poor; 0.21-0.40 fair; 0.41-0.60 moderate; 0.61-0.80 good; 0.81-1.00 very good agreement). The agreement was moderate for determining the T class (κ=0.45, overall agreement 58%), poor for the N class (κ=0.13, 42%) and fair for the M class (κ=0.22, 58%). The agreement for overall staging of NSCLC was poor (κ=0.20, 45%). The major source of disagreement was that metastases were present more frequently and/or in larger number on 18F-FDG PET/CT than MDCT in the contralateral mediastinal, supraclavicular, and distant lymph nodes, as well as in the bones and suprarenal glands. Since 18F-FDG PET/CT detected more regional and distant metastases than MDCT, we conclude that FDG PET/CT is useful for staging/restaging and planning treatment of patients with NSCLC. KEYWORDS: Non-small cell lung cancer, positron emission tomography, multidetector computed tomography, metastases detection.

Evaluation of the SIOPEN semi-quantitative scoring system in planar simpatico-adrenal MIBG scintigraphy in children with neuroblastoma.

Neuroblastoma is the most common malignancy in children comprising 7.6% of all infantile cancers. MIBG scintigraphy is a mandatory neuroblastoma diagnostic test, which is among others methods, semi-quantified by the SIOPEN method. The aim of this study was to test both the skeletal and the soft tissue segments of the SIOPEN scoring method in the diagnostic milieu and to correlate them with the Curie score. Since there is little knowledge of their diagnostic power, the following variables were tested: VMA, HVA, LDH, and MYCN, ferritin, bone marrow infiltration, the INSS and the INPC classification. The cross-sectional study with repeated measurements of 143 scintigrams was performed on 76 pediatric patients with suspected or proven neuroblastoma, who had been referred to the Center for Nuclear Medicine of the Clinical Center of Serbia in the period 2007-2012. The range of the SIOPEN soft tissue scores was 0-5. The range of the SIOPEN skeletal scores was 0-57. The range of the Curie scores was 0-26. The skeletal SIOPEN scores were significantly higher in bone marrow positive children, in children with pathologically elevated urinary VMA levels and in children having a more advanced clinical stage. There was no difference in the SIOPEN soft tissue score due to higher VMA levels, or depending on the clinical stage and positive bone marrow assessment. There was no difference between the SIOPEN skeletal and soft tissue scores on one hand and the histological grade of the tumor; elevated or normal levels of HVA, LDH, NSE and ferritin, or the presence or absence of MYNC amplification in the neuroblastoma cell line, on the other hand. The results of both SIOPEN scores showed a high linear correlation with the Curie score. The conclusion is that the soft tissue segment of the SIOPEN score needs further elucidation in a more controlled milieu. Excellent correlation between all segments of the two semi-quantitative scoring methods speaks in favor of the application of the complete SIOPEN scoring system in every day mIBG scanning.

Utility of MRI in detection of PET-CT proven local recurrence of pancreatic adenocarcinoma after surgery.

The aim of this prospective study was to investigate the accuracy and inter-observer reliability of MRI in detection of local recurrence (LR) of pancreatic adenocarcinoma (PAC) after surgery, which was proved by PET-CT and access correlation between functional MRI and PET parameters. Forty-five patients who underwent PET-CT and MRI for follow-up purposes after radical operation of PAC were included. Twenty-three were PET positive (study group) and 22 negative for LR (control group). MR examination was performed within one month after PET-CT and three readers who were blind for PET-CT findings searched LR in T2W, 3D-dynamic post-contrast T1W-FS and DWI sequences, respectively. Sensitivity and specificity were calculated while inter-reader agreement was estimated by Cronbach's Alpha reliability coefficient (CARC). Apparent diffusion coefficient (ADC) of LR was correlated with the size (maximal diameter) and functional PET-CT parameters: mean and maximum standardized uptake values (SUVmean, SUVmax), metabolic tumor volume (MTV) and total lesion glycolysis (TLG), using Spearman's correlation coefficient (rS). Sensitivity and specificity among three readers in detecting the LR were 70% and 77-84% in T2W (CARC 0.806), 91-100% and 100% in 3D post-contrast T1W-FS (CARC 0.980), and both 100% in DWI sequences (CARC 1.000). Moderate inverse correlation was found between the ADC and SUVmean (rS = - 0.484), MTV (rS = - 0.494), TLG (rS = - 0.519) and lesion size (rS = - 0.567). MRI with DWI shows high diagnostic accuracy in detecting the LR of PAC in comparison to PET-CT as reference standard. ADC significantly inversely correlates with standard and advanced PET parameters and size of LR.

CT volumetry of normal pancreas: correlation with the pancreatic diameters measurable by the cross-sectional imaging, and relationship with the gender, age, and body constitution.

PURPOSE: The purpose of this study was to estimate the volume of normal pancreas in adults using the CT volumetry (summation of the areas technique), analyze the correlation between the volume and the diameters of pancreas, which are measurable by the cross-sectional imaging, and assess the relationship with the gender, age, and body constitution. METHODS: 220 CT examinations were analyzed retrospectively (102 females, 118 males; age 16-82, average 56). Following diameters were measured: cranial-caudal-CC(pancreas), CC(body&tail), CC(body), CC(head); anterior-posterior--AP(tail), AP(body), AP(head); lengths--LL(head), L(body&tail); and maximal transversal diameter of the L1 vertebral body (LL(L1)) and thickness of the abdominal subcutaneous fat (AP(ASF)), as markers of body constitution. RESULTS: The average volume of the pancreas was 79.2 ± 24.1 cm3 (ranging from 37.4 to 168.2 cm3). Pancreatic volume strongly correlated with all measured diameters of the pancreas (P < 0.0001). Pancreatic volume significantly correlated with gender (M:F = 86.1:72.8 cm3, P = 0.002) and the LL(L1) (r = 0.185, P = 0.008), and did not correlate with the age (r = -0.110, P = 0.151) and the AP(ASF) (r = -0.115, P = 0.104). Correlation of vertebral body-pancreas volume ratio of each subject and the age was strongly negative (r = -0.202, P = 0.006). CONCLUSIONS: Marked individual variations in normal pancreas volume were observed. Pancreatic volume could be computed using the diameters measurable by the cross-sectional imaging employing the formula: V = (AP(tail) + AP(body))/2 × L(body&tail) × CC(body) + (AP(head)/2)2 × 3.14 × CC(head).

The clinical value of scintigraphy of neuroendocrine tumors using (99m)Tc-HYNIC-TOC.

PURPOSE: To assess the value of whole body scintigraphy using (99m)Tc-HYNIC-TOC (Tektrotyd) and with single photon emission computerized tomography (SPECT) in the detection of primary and metastatic neuroendocrine tumors (NETs). METHODS: Thirty patients with different neuroendocrine tumors, mainly gastroenteropancreatic (GEP), were investigated. Whole body scintigraphy was performed 2 h (if necessary 10 min and 24h) after i.v. administration of 740 Mbq (99m)Tc-Tektrotyd, Polatom. In cases of unclear findings obtained by whole body scintigraphy, investigation was followed by SPECT. RESULTS: From 12 patients with NETs of unknown origin, there were 10 true positive (TP), and 2 false negative (FN) findings. Diagnosis was made with SPECT in 6 patients. From 8 patients with gut carcinoids, there were 4 TP, 2 true negative (TN), one FN, and one false positive (FP) finding. Diagnosis was made with SPECT in 2 patients. From 7 patients with neuroendocrine pancreatic carcinomas there were 4 TP and 3 TN findings. Diagnosis was made with SPECT in 2 patients. From 3 patients with gastrinomas there were 2 TP findings and one TN findings. Diagnosis was made with SPECT findings in 2 patients. Sensitivity of (99m)Tc-HYNIC-TOC was 87%, specificity 86%, positive predictive value 95%, negative predictive value 67% and accuracy 87%. CONCLUSION: We concluded that scintigraphy with (99m)Tc-Tektrotyd is an useful method for diagnosis, staging and follow up of the patients with NETs.

Scintigraphic detection of colon carcinomas with iodinated monoclonal antibodies.

PURPOSE: The aim of this study was to evaluate the clinical reliability of the immunoscintigraphy with iodinated monoclonal antibodies for the detection of metastases and recurrences of colon carcinomas. METHODS: A total of 45 patients with colon carcinoma was investigated with gamma camera, after intravenous application of iodinated monoclonal antibodies. RESULTS: The sensitivity of the method was 90%, specificity 86%, positive predictive value 93%, negative predictive value 80% and accuracy 87%. There was statistically significant relationship between immunoscintigraphic and ultrasonographic (US) findings (p=0.005). Also, there was significant relationship between immunoscintigraphy and Dukes stage (p=0.019). Tumor marker levels were not significantly correlated with immunoscintigraphic findings (p<0.05). Significant difference was noted in patients with positive findings for malignancy on US and immunoscintigraphic findings (p=0.006), i.e. patients with positive findings for malignancy had more frequently immunoscintigraphic findings of malignancy. Correlation with other diagnostic procedures (rectoscopy, colonoscopy, CT) did not show significant correlations. CONCLUSION: We conclude that immunoscintigraphy can be helpful in the detection of metastases and recurrences of colon carcinomas.

The utility of 99mTc-EDDA/HYNIC-TOC scintigraphy for assessment of lung lesions in patients with neuroendocrine tumors.

UNLABELLED: Our aim was to assess clinical utility of 99mTc-EDDA/HYNIC-TOC scintigraphy for evaluation of lung lesions in patients with neuroendocrine tumors (NETs). Single photon emission computed tomography (SPECT) of the thorax and whole body scintigraphy were performed in 34 patients using 99mTc-EDDA/HYNIC-TOC. Visual assessment was complemented by semiquantitative evaluation based on tumor to non-tumor (T/NT) ratio. Clinical, laboratory, and histological findings served as the standard for comparison. Enhanced tracer uptake was observed on both SPECT and whole body scintigraphy in 29 of 34 patients (88% sensitivity). T/NT ratios were significantly higher on SPECT than whole body images (2.96+/-1.07 vs.1.70+/-0.43, p KEYWORDS: 99mTc-EDDA/Hynic-TOC, lung involvement of NETs, T/NT ratio.

Is there a future role for immunoscintigraphy in the diagnosis of colorectal carcinoma?

Immunoscintigraphy combines the advances in immunology and nuclear medicine to target tumor sites. Visualization of colorectal carcinomas is based on different monoclonal antibodies and their fragments against tumor-associated antigens labeled with gamma emitting radionuclides which accumulate in the tumor tissue due to their interaction with corresponding antigens. Available data on the role of immunoscintigraphy in detection of recurrence and metastases of colorectal carcinomas are reviewed. Despite a variety of investigations related to the application of immunoscintigraphy in diagnostics of colorectal cancer, using different radiolabeled immunoreactive agents and imaging methods there has not been a consensus among the investigators regarding the best modality of the method, including the specific radiopharmaceutical for this purpose. Some general conclusions concerning potentials of immunoscintigraphy in such diagnostics, including expectancy of the newly developed SPECT/CT systems, are suggested. The possibilities of PET imaging of colorectal carcinomas using monoclonal antibodies labeled with positron emitting radionuclides, as well as of the radioimmunoguided surgery and radioimmunotherapy are also discussed. Key words: Immunoscintigraphy, colorectal cancer, recurrences, metastases, radioimmunoguided surgery, SPECT/CT.

The accuracy of ultrasonography in classification of groin hernias according to the criteria of the unified classification system.

BACKGROUND: The modern concept of type-related individualized groin hernia surgery imposes a demand for precise and accurate preoperative determination of the type of groin hernia. The aim of this prospective study was to evaluate the accuracy of ultrasonography in classification of groin hernias, according to the criteria of the unified classification system. Unified classification divides groin hernias into nine types (grades): type I (indirect, small), II (indirect, medium), III (indirect, large), IV (direct, small), V (direct, medium), VI (direct, large), VII (combined-pantaloon), VIII (femoral), and O (other). PATIENTS AND METHODS: One hundred and twenty-five adult patients with clinically diagnosed or suspected groin hernias were examined. Ultrasonography of both groins was performed with a 5 to 10-MHz linear-array transducer. Preoperative ultrasonographic findings of type of groin hernia were compared with the intraoperative findings, which were considered the gold standard. RESULTS: Total accuracy of ultrasonography in determination of type of groin hernia was 96% (119 of 124 correct predictions of type of groin hernia compared with surgical explorations). All hernias of types I, IV, V, VII, and VIII were correctly identified with ultrasonography (sensitivity and specificity 100%). In the remaining five cases of the 124 (4%), hernia was incorrectly classified with ultrasonography: type VI (direct, large) was misdiagnosed as type III (indirect, large) in three cases, type III as type VI in one case, and type III as type II (indirect, medium) in one case. The sensitivity and the specificity of ultrasonography in classifying type II were 100 and 99%, respectively, for type III, 85 and 97%, and for type VI, 90 and 99%. CONCLUSION: Ultrasonography of the groin regions could be used with great accuracy for precise classification of groin hernias in adults. Each type of groin hernia, according to the unified classification system that we used for classification, has a characteristic ultrasonographic presentation, which is demonstrated in this study.

Radioimmunoscintigraphy of colorectal carcinomas with three different radiopharmaceuticals.

The aim of the study was evaluation of the clinical reliability of the immunoscintigraphy for the detection of metastases and recurrences of colorectal carcinomas using three different radiopharmaceutical substances. With IMACIS 1, the number of true negative findings (TN) was 4/7 and true positive (TP) 3/7, while in one patient, the results of immunoscintigraphy significantly influenced the therapeutical management. With INDIMACIS 19-9, there were 2/8 TN and 6/8 TP. In three patients, immunoscintigraphy results influenced patient further management. With ONCOSCINT in 2 patients findings were TN, in one FN and in one FP. In 3 patients, immunoscintigraphy influenced the management of the patient. Other imaging methods (CT, US, MRI) have advantage in detection of liver metastases, while immunoscintigraphy is more specific for the assessment of reccurences of the abdominal tumors. Thus immunoscintigraphy should be applied in patients with suggested recurrences and inconclusive outcome of routine diagnostic workup.

Standardized perfusion value of the esophageal carcinoma and its correlation with quantitative CT perfusion parameter values.

PURPOSE: Standardized perfusion value (SPV) is a universal indicator of tissue perfusion, normalized to the whole-body perfusion, which was proposed to simplify, unify and allow the interchangeability among the perfusion measurements and comparison between the tumor perfusion and metabolism. The aims of our study were to assess the standardized perfusion value (SPV) of the esophageal carcinoma, and its correlation with quantitative CT perfusion measurements: blood flow (BF), blood volume (BV), mean transit time (MTT) and permeability surface area product (PS) of the same tumor volume samples, which were obtained by deconvolution-based CT perfusion analysis. METHODS: Forty CT perfusion studies of the esophageal cancer were analyzed, using the commercial deconvolution-based CT perfusion software (Perfusion 3.0, GE Healthcare). The SPV of the esophageal tumor and neighboring skeletal muscle were correlated with the corresponding mean tumor and muscle quantitative CT perfusion parameter values, using Spearman's rank correlation coefficient (rS). RESULTS: Median SPV of the esophageal carcinoma (7.1; range: 2.8-13.4) significantly differed from the SPV of the skeletal muscle (median: 1.0; range: 0.4-2.4), (Z=-5.511, p<0.001). The cut-off value of the SPV of 2.5 enabled discrimination of esophageal cancer from the skeletal muscle with sensitivity and specificity of 100%. SPV of the esophageal carcinoma significantly correlated with corresponding tumor BF (rS=0.484, p=0.002), BV (rS=0.637, p<0.001) and PS (rS=0.432, p=0.005), and SPV of the skeletal muscle significantly correlated with corresponding muscle BF (rS=0.573, p<0.001), BV (rS=0.849, p<0.001) and PS (rS=0.761, p<0.001). CONCLUSIONS: We presented a database of the SPV for the esophageal cancer and proved that SPV of the esophageal neoplasm significantly differs from the SPV of the skeletal muscle, which represented a sample of healthy tissue. The SPV was validated against quantitative CT perfusion measurements and statistically significant correlation was proved.

Assessment of early epirubicin cardiotoxicity in women with breast cancer.

BACKGROUND: Anthracycline-containing chemotherapy has been the most frequently observed cause of iatrogenic cardiac damage in patients with breast cancer. The purpose of this study was to evaluate by radionuclide ventriculography whether a mean cumulative dose of epirubicin (MCDE) induced left ventricular (LV) systolic or/and diastolic dysfunction in 32 patients treated for breast cancer. MATERIALS AND METHODS: Thirty-two patients with breast cancer according to chemotherapeutical trials received MCDE of 360 m/m2. All patients were studied before and after they completed chemotherapy with radionuclide ventriculography at rest. Systolic and diastolic left ventricular parameters were assessed. RESULTS: Diastolic left ventricular parameters and R-R intervals significantly differed before and after completed chemotherapy in our patients: peak filling rate (PFR: 2.8 +/- 0.6 vs. 2.2 +/- 0.7 EDV/sec) and time to filling rate (TPFR: 182 +/- 48 vs. 2.25 +/- 50 msec), R-R: 723 +/- 51 vs. 620 +/- 45 msec, p < 0.01. Ejection fraction, as a systolic parameter/did not significantly differ before and after completed chemotherapy (EF: 59 +/- 7 vs. 58 +/- 6%), p > 0.05. CONCLUSION: Our results indicate that left ventricular diastolic dysfunction even at an MCDE of 360 mg/m2 may be an early sign of epirubicin cardiotoxicity.

A new radionuclide approach for the quantification of left ventricular volumes: the 'geometric count based' method.

A new radionuclide method, called the 'geometric count based' (GCB) method, has been developed for the quantification of absolute left ventricular volume. As the method is based on planar radionuclide ventriculography, it is non-invasive and simple, and avoids the relatively cumbersome and longer lasting, dynamic procedure using single photon computed emission tomography, which can be used for achieving the same goal. The purpose of this study was to describe the exactness of the theoretical approach to the method and validate its accuracy both by physical experiments and the initial clinical trial, as compared to contrast ventriculography. Count based data were combined with the geometric based data assuming an ellipsoid left ventricular shape with identical short axes. The following equation for computing left ventricular end diastolic volume, EDV (in ml) was developed: EDV=2cMCtot/Cmax, where c is the manually drawn short axis (one row pixel ROI) of the prolate ellipsoid in LAO 45 degrees (cm), M is the calibrated pixel size (in cm2), Ctot is the total counts in LV ROI, and Cmax is the maximum pixel counts in the LV ROI. Physical experiments with two different 'heart shaped' phantoms were used to compare the results obtained by the GCB method with the true phantom volumes and with the method assuming LV ball shape (BLV), developed by other authors. The true volumes of cylindrical and ellipsoid phantoms of 112.5 ml and 190.5 ml were computed to be 114 ml and 196 ml by the GCB and 168 ml and 180 ml by the BLV methods, respectively. In a clinical study, GCB volumes were compared to volumes measured by using single plane contrast ventriculography in 38 coronary patients. A good correlation between the GCB method and contrast ventriculography was obtained both for EDV and end systolic ventricular volumes (r=0.94, r=0.90). Both phantom and initial clinical studies indicate that the GCB method is an accurate, non-invasive and simple radionuclide method for measuring left ventricular volumes. Additionally, it could be used even in the smallest nuclear medicine units, for example in intensive care units where there are mobile cameras.

Prognostic value of myocardial viability determined by a 201Tl SPECT study in patients with previous myocardial infarction and mild-to-moderate myocardial dysfunction.

The prognostic meaning of myocardial viability is most important in patients with severe left ventricular dysfunction and ischaemic heart disease, but its prognostic significance in patients with previous myocardial infarction and mild-to-moderate myocardial dysfunction is uncertain. The aim of this study was to assess the prognostic value of a 201Tl single photon emission computed tomography (SPECT) rest-redistribution study in patients with previous myocardial infarction, ischaemic heart disease and mild-to-moderate myocardial dysfunction. Myocardial viability was assessed in 55 patients (50 male; mean age 58+/-9 years) by 201Tl SPECT rest-redistribution (after 4 h) scintigraphy. All patients had previous myocardial infarction (>3 months) and angiographically documented coronary artery disease, with the mean ejection fraction of 43+/-10%. Out of 55 patients, 20 were medically treated and 35 were revascularized. The follow-up period for adverse cardiac events, including death and non-fatal myocardial infarction, was 12 months. 201Tl SPECT study was positive for myocardial viability in 36 patients (65%) and negative in 19 patients (35%). Sensitivity, specificity, positive and negative predictive values for functional improvement in the follow-up period were 85%, 75%, 92% and 60%. Out of seven (13%) cardiac events in the follow-up period (four cardiac deaths and three reinfarctions), five occurred in 20 medically treated patients and two in 35 revascularized patients (25% vs 6%, P <0.05). Absence of myocardial viability was the only variable associated with adverse cardiac events (P =0.02). Survival at 12 months, as determined by using Kaplan-Meier analysis, was 56% for medically treated and non-viable patients, 80% for revascularized and non-viable patients, 91% for medically treated and viable patients, and 100% for revascularized and viable patients (P =0.0034). These findings suggest that in patients with previous myocardial infarction and mild-to-moderate myocardial dysfunction, the absence of myocardial viability as determined by the 201Tl SPECT study was the only variable associated with adverse cardiac events. The best 12 month survival was observed in revascularized viable patients, whereas the worse prognosis was found in non-viable, medically treated patients.