Intraoperative cytodiagnosis of ovarian tumors

The authors report on their experience in applying rapid intraoperative cytologic analysis in 83 benign and 54 malignant ovarian tumors of various pathohistologic types. The entire procedure, from the receiving of the material, fixing and staining, up to the cytologic diagnosis, did not last longer than 6 minutes. The cytologic findings were negative in all benign tumors except one (99.3%). In cases of malignant tumors the cytologic diagnosis was correct in 49 cases (96.4%). On the basis of cytomorphologic criteria it was also possible to predict the pathohistologic type of most of investigated tumors. The high overall accuracy rate of intraoperative cytologic examination permits the surgeon to make a quick and adequate therapeutic decision concerning the extent of intervention.     

The accuracy of mammography: an analysis of 655 histologically verified cases

The accuracy of the mammography was studied on 655 breasts which were investigated and biopsied or treated surgically shortly afterward. The mammography reports were classified according to the confidence level of the statements. In 279 cases the mammography was interpreted as positive for malignancy. A total of 224 malignant tumors were found. In 30 cases the malignant lesion was missed on the mammograms. The diagnoses made with a high degree of certainty were correct in 90% of the cases. The sensitivity of the mammography was 86.6% at a specificity of 80.7% for malignant tumors. The value of mammography for the diagnosis of chronic mastopathy could not be determined objectively. The sensitivity for solitary benign tumors or cysts was 80%.     

The value of palpation as basis for decision on lymph node excision

The purpose of the study was to evaluate palpation of the regional lymph nodes in control examinations of patients with malignant disease. A retrospective review of the medical records of 188 cases in which the patients had had an extirpation of the regional lymph nodes was performed. We have compared the preoperative findings through palpation with the histological diagnosis. The patients were grouped according to the region in which the lymph node removal had been done. The specificity of palpation when the histological diagnosis was malignant was (with 95% confidence limits), in the axilla 0.65 (0.54-0.75), in the inguinal region 0.86 (0.75-0.94) on the neck 0.83 (0.52-0.98) and in the suprahyoid region 0.58 (0.28-0.85). In conclusion, palpation of regional lymph nodes is not a sufficient control examination in the estimation of the course of malignant disease. Supplementary examination methods are recommended.     

Ultrasonographic diagnosis of the parotid gland tumors--experience with 310 patients

The diagnostic value of ultrasonography of parotid gland tumors is to determine if the tumor is benign or malignant. In the period from 1984 to 1995, ultrasound examinations were performed on 310 patients with a space-occupying lesion of the parotid gland, using real time 3.75 MHz, 5 MHz and 7.5 MHz transducers. From a histological standpoint, there were 246 benign tumors, namely 144 pleomorphic adenomas, 35 adenolymphomas, 23 other adenomas, 14 cysts, 9 neurinomas, 3 lymphomas, 9 hemangiomas, 3 lymphangiomas and 6 lymph nodes. Another 64 were malignant tumors, namely, 7 mucoepidermoid carcinomas, 7 acinic cell carcinomas, 7 adenoid cystic carcinomas, 15 adenocarcinomas, 7 squamous cell carcinomas, 4 undifferentiated carcinomas, 10 carcinomas in pleomorphic adenoma, 3 malignant lymphomas, 2 metastatic carcinomas and 2 other carcinomas. According to our criteria for ultrasonographic diagnosis of the parotid gland tumors, the benign or malignant pattern was determined by the following findings; shape, boundary echo, internal echo and posterior echo. But in some cases we could not differentiate the two echogram patterns, and we call them the intermediate pattern. The total accuracy rate of the diagnosis of the 310 patients was 79.0%. (The intermediated pattern was found in 39 patients and these cases were considered to be misinterpreted.) In 268 primary parotid gland tumors (210 benign and 58 malignant tumors) excluding recurrent cases, lymph nodes, hemangiomas and so on, the sensitivity was 62.1%, the specificity was 91.4% and the total accuracy rate was 85.1%. The accuracy of sonography in determining whether a tumor was benign or malignant was affected by the tumor size; tumors smaller than 2 cm in diameter showed a tendency to be diagnosed as benign and tumors larger than 6 cm in diameter as malignant. In the 268 cases, the importance of the pattern of the boundary echo and the internal echo in differentiating benign and malignant tumors was examined retrospectively. In the 210 primary benign tumors, the benignity of the tumors was assessed correctly in 84.3% by the boundary echo and 85.7% by the internal echo. In the 58 primary malignant tumors, the malignancy was assessed correctly in 58.6% by the boundary echo and 43.1% by the internal echo. Strong echos, which are the signs of a malignant pattern, were seen in 14.8% of the benign tumors, and in 51.7% of the malignant tumors. It can be concluded that ultrasonography is very reliable for the examination of parotid gland tumors.     

Opisthorchiasis control in northeast Thailand: proposal for a new approach

In a prospective study 102 patients with malignant tumors of the tongue, oropharynx, hypopharynx, and larynx were staged by means of MRI. Special attention was directed at those tumour extensions that could influence the treatment strategy. The MR findings were correlated with the results of palpation, endoscopy, ultrasound, computerised tomography and histopathological findings. MRI showed a good delineation of pT2 to pT4 tumours. However, visualisation of small tumours in the soft palate, the tonsils, the pyriform sinuses and vocal cords was difficult. In conclusion our results suggest that in addition to endoscopy MRI is a valuable tool for the examination of tumours of tongue, oropharynx, hypopharynx and larynx.     

The value of intraoperative consultation (frozen section) in the diagnosis of ovarian neoplasms

BACKGROUND: Frozen Section is an important diagnostic tool to determine the nature of ovarian masses. However, like other diagnostic tools, frozen section also has some pitfalls. We aimed to discuss the source and the nature of inaccuracies associated with this procedure. METHODS: In this retrospective study 360 cases of ovarian masses examined by frozen section were re-evaluated. The sensitivity, specificity and predictive values of frozen section diagnosis of ovarian tumors were calculated. The reasons for the erroneous frozen-section diagnoses were discussed. RESULTS: Overall diagnostic agreement for ovarian lesions was 94.2%. Disagreements were found in nine cases (2.5%). Diagnosis was deferred to permanent sections in 12 cases (3.3%). The sensitivity for malignant tumors was 93.1% and specificity was 99.2%. The sensitivity for benign tumors was 99.2% and specificity was 92.1%. Most problematic cases were mucinous tumors, followed by tumors resembling fibrothecomas, in addition sections without viable tissue fragments or presence of extensive hemorrhage and necrosis also obscured the frozen diagnosis. Another factor was the lack of an effective communication between the surgeon and the pathologist. CONCLUSIONS: For an effective usage of this method not only the pathologist but also the surgeons must know the pitfalls of this method and also there must be good communication between the pathologist and the surgeon. Especially deferred cases should be minimized by good communication. In fact it's an intraoperative consultation method that enables the pathologist to gather all the preoperative, intraoperative findings and to be familiar with the further treatment plan of the patient.     

Preoperative and intraoperative topographic diagnosis of insulinomas

Altogether 120 patients with organic hyperinsulinism underwent clinical examination and treatment (38 male, 82 female, mean age 44.2 +/- 4.6 years). The cause of hyperinsulinism was benign insulinomas in 96 (80.0%), malignant tumors in 9 (7.5%), and hyperplasia of beta cells in 6 (5.0%). In 9 (7.5%) patients the origin of hyperinsulinism was not diagnosed. The tumor was localized in the head, body, and tail of the pancreas in 31.8%, 36.4%, and 31.8% of cases, respectively. Intraoperative ultrasonography (IOUS) was undertaken in 37 patients, and in 83 cases only intraoperative palpation was done. Arterial stimulated venous sampling (ASVS) was performed in 17 patients (blood was sampled from the right hepatic vein for determination of the insulin level after arterial stimulation by calcium gluconate in different parts of the pancreas). The sensitivity of ultrasonography (US) was 29.5%, computed tomography (CT) 24.2%, angiography 55.9%, superselective angiography (branches of the celiac trunk) 72.2%, and intraoperative palpation 90.0%. ASVS showed an accuracy of 90.0%. Combining angiography with ASVS gave an exact diagnosis of hyperinsulinism in 100% of cases, and IOUS revealed tumors in 100% of cases. Hyperplasia of beta-cells was diagnosed only by means of ASVS. A total of 117 patients underwent surgery, including distal resection of pancreas (n = 39), enucleation of tumor (n = 70), and laparotomy (n = 8). The postoperative mortality associated with insulinomas was 7.7%. The frequency of postoperative complications was 43.6%. Benign insulinomas recurred at a rate of 5.4%. Patients with malignant insulinomas had a 5-year survival of 66.0%. The diagnosis of insulinomas was achieved by a combination of selective angiography, ASVS, and IOUS.     

Does autopsy of antenatally diagnosed malformed foetuses aid genetic counselling?

BACKGROUND: Many pregnancies are terminated because of ultrasonographic diagnosis of malformation in the foetus. A detailed foetal autopsy is needed to arrive at a definite diagnosis on the basis of which genetic counselling can be provided. METHODS: Sixty-one foetuses, terminated because of antenatal diagnosis of congenital malformations by ultrasound, were autopsied. The ultrasound diagnosis was compared with the diagnosis reached after autopsy. RESULTS: In 31 cases (51%) the autopsy provided additional findings. In 21 cases (34.4%), the autopsy changed the primary diagnosis. The revised diagnosis led to a change in the risk of recurrence in 18 cases (29.5%). CONCLUSION: Genetic counselling depending solely on ultrasonographic foetal diagnosis may be erroneous. For appropriate genetic counselling, a detailed foetal examination should be carried out after termination in cases with ultrasonographically detected congenital malformations.     

The role of clinical autopsy in monitoring the quality of the clinical diagnosis in an emergency department

OBJECTIVE: A total of 61 autopsies performed in patients died in emergency department of a university hospital were retrospectively analysed and the findings were compared with clinical diagnoses. METHODS: Sensitivity and specificity of the clinical diagnoses and the correction of medical procedures were measured. The influence of age and sex of patients was analyzed using Fisher's exact test and chi-square-test. RESULTS: The most common causes of death were cardiovascular diseases (52.46%). Autopsy showed unexpected major findings in 44.26% of cases. Major discrepancies between the autopsy reports and the clinical diagnoses, were present in 26.22% of all cases. Absolute concordance between clinical and autopsy diagnoses was obtained in 44.26% of cases. The major sensitivity of clinical diagnosis was found in cerebrovascular disorders (100%), upper digestive hemorrhage (100%), and acute myocardial infarction (82.35%). The lowest sensitivity was found in malignant tumors (16.66%), hemorrhagic pancreatitis (0%) and bowel infarction (0%). The patient cares were correct in 68.85% of cases. No statistically significant differences were observed in relation to age and sex. CONCLUSIONS: We concluded that autopsy is a useful method for evaluate diagnostic procedures and quality of medical cares in emergency departments.     

Value of bronchoscopic pneumonia diagnosis: prospective study

The frequent detection of benign liver lesions during ultrasound routine examination and a possible curative therapy of early detected malignant tumors require a reliable method of differentiation. Conventional gray-scale ultrasound, according to this problem, has been extended by the Duplex technique and color Doppler ultrasound. Measurement of blood flow velocity by Doppler in the center and at the periphery of liver lesions is not reliable enough to distinguish between benign and malignant lesions. Color Doppler ultrasound possesses some reliable criteria for differentiation. A central spot could be detected in 2 out of 12 hemangiomas, a giant spot in 1 out of 3 giant cavernous hemangiomas. The halo sign without detectable blood flow is considered to be specific for malignancy. We found this sign in 26 out of 81 malignant liver lesions and only in one benign lesion (sensitivity 32%, specificity 97%). The vascularization of focal liver lesions is excellently demonstrated with color Doppler ultrasound. This is very helpful for the diagnosis (e.g. "chaotic blood vessel architecture' in malignant tumors) and for the therapy of focal liver lesions (e.g. follow-up examinations after chemotherapy or chemoembolization). Therefore, application of Duplex and color Doppler ultrasound is highly recommended as a noninvasive diagnostic method of first choice for unknown liver lesions.     

Reconstructive operations on the biliary tract

Fifty cases in which reconstruction of the biliary system was carried out were reviewed. In 25 cases the operation was done during the treatment of malignant neoplasms. The other 25 patients were treated for benign conditions. Delayed stricture of the biliary anastomosis occurs more frequently following operation for benign post-traumatic obstruction than following reconstruction for other conditions. This is probably a result of: (1) greater regional scarring, (2) local infection, and (3) technical imperfections in the reconstituted biliary anastomosis. Certain primary malignant tumors may be difficult to recognize by both gross and microscopic examination. In six cases of biliary obstruction resulting from malignant neoplasms in the present series, exploration had been carried out some time previously, and in four of them an erroneous diagnosis of benign biliary obstruction was made.End-to-end anastomosis of the duct above and below the point of obstruction is the method preferred in the treatment of benign biliary stricture. Intrahepatic and extrahepatic biliary-enteric anastomoses have been used successfully in selected cases.     

Pancreatic cystic lesions of difficult diagnosis

Cystic lesions of the pancreas include pseudocysts (90%), cystic tumors and true cysts. Preoperative diagnosis will guide the therapeutic strategy. Certain malignant or potentially malignant cysts (mucinous cystadenomas and cystadenocarcinomas) should be removed surgically while others (pseudocysts and serous cystadenomas) are almost always benign. In 3 out of 4 cases, the clinical setting guides diagnosis and with adequate imaging (sonography, computed tomography, endosonography) the correct diagnosis can be obtained. Inversely, the nature of a unique macrocyst it is often difficult to identify. Puncture is useful to determine tumor markers which are sometimes helpful in complementing information provided by imaging techniques.     

Demyelinating disease versus tumor in surgical neuropathology. Clues to a correct pathological diagnosis

Clinical presentations as well as radiological and histopathological findings in biopsies from patients with multiple sclerosis (MS) or other demyelinating disorders of the central nervous system are sometimes misleading, resulting in an erroneous diagnosis of brain or spinal cord tumor. We report 17 patients who presented with symptoms mimicking those of brain (14 cases) or spinal cord (three cases) tumors. Computerized tomography or magnetic resonance imaging studies or both were interpreted as consistent with a tumor in each case. All patients underwent surgery, and all 17 pathological specimens were eventually diagnosed as showing demyelinating disease, usually consistent with MS. In each case we examined a variety of histological features and immunohistochemical studies and addressed their relative importance in considering the diagnosis of MS. All cases showed perivascular lymphocytic inflammation with variable amounts of macrophage infiltration, necrosis, and edema. The hypercellularity of the lesions and the presence of atypical reactive astrocytes with mitotic figures were the disturbing features that might have led to the erroneous diagnosis of an astrocytic neoplasm. Immunohistochemistry for astrocytic (glial fibrillary acidic protein) and macrophage (HAM-56) markers are helpful in evaluating biopsies. Our results emphasize the need to perform special stains (i.e., for myelin and axons) that demonstrate myelin loss and relative preservation of axons and allow a correct diagnosis.     

Imaging of polypoid endoluminal growing bile duct tumors

We reviewed the radiological documents and protocols of 196 cases of bile duct tumors examined over a period of 12 years: 20 of them (10.2%) presented with a polypoid endoluminal growth. The aim of this study was to provide a better knowledge about the radiological features of this less frequent kind of tumor. In these 20 cases, the correct diagnosis of bile duct lesion was provided in 100% by E.R.C.P., P.T.C., U.S. as well as C.T., and in 42% by arteriography. The correct diagnosis of tumor was made by E.R.C.P. in 86%, by P.T.C. in 88%, by U.S. in 61%, by C.T. in 63%, and by arteriography in 25%. A correct diagnosis of tumor could be reached in all cases by combination of several examination techniques. The anatomopathological diagnosis was: bile duct adenocarcinoma (7 cases), adenocarcinoma of the ampulla of Vater (4), villous adenoma of Vater's ampulla (2), cystadenoma (1), cystadenocarcinoma (1), hepatocellular carcinoma (1), apudoma (1), and metastases (3).     

Differential diagnosis of cystic adnexa processes. A comparison of the value of sonography, puncture cytology and color and biochemical biopsy analysis with histology

The histological diagnoses of 109 cystic adnexal masses were compared with the results of ultrasound examination, color, cytology and biochemical analyses of cystic fluid. In cases with suspicious results by ultrasound examination most of the malignant tumors but only a small number of benign cystic masses were found. Macroscopic and cytologic examinations gave no further informations. By biochemical analyses functional cysts could be detected with 86% of sensitivity and 100% of specificity by a combination of estradiol, progesterone and Ca 12-5. Endometrioma and malignant tumors could be separated from other adnexal masses (specificity 95%, sensitivity 100%) with a combination of Tag 72-4 and CASA.     

Mitochondrial neurogastrointestinal encephalomyopathy presenting with protein-losing gastroenteropathy and serum copper deficiency: a case report

OBJECTIVE: To evaluate ultrasonic and magnetic resonance imaging (MRI) and Doppler examination in fetal sacrococcygeal teratoma (SCT), in respect to the postnatal findings and histological type of the tumor. STUDY DESIGN: Nine pregnancies complicated by histologically mature fetal sacrococcygeal teratoma in four cases and by immature/malignant teratoma in five cases. Transabdominal ultrasonic imaging and Doppler velocity waveforms were recorded during the second or last trimester in all cases, first trimester ultrasound examination was carried out in six cases and last trimester MRI in five cases. These findings were compared with postnatal and operative findings of the children. RESULTS: Ultrasound examination did not reveal intrapelvic parts of SCT, but this was possible by MRI. Velocity waveforms of the tumor arteries were similar in all histological types and the resistance index varied from 60 to 70. The mean gestational age at antepartal diagnosis was 25.2 weeks. Large tumor size with relatively large proportion of solid components was often recognized in cases with malignant/immature histology. CONCLUSIONS: Antepartal MRI is useful for examination of fetal SCT, but reliable differentiation of mature and immature SCT is not possible antepartally.     

Accuracy of endoscopic ultrasonography in the diagnosis and staging of gastric cancer and lymphoma

BACKGROUND: There is a need to assess the sensitivity, specificity, and predictive value of endoscopic ultrasonography (EUS) in the diagnosis and staging of gastric cancer and lymphoma. METHODS: A prospective study was performed on 86 patients with endoscopic gross appearance suspicious for cancer or lymphoma. Biopsies with endoscopic forceps were always carried out before EUS. All patients underwent laparotomy for final diagnosis, staging, and eventually treatment. The results of EUS were correlated with the histologic findings of the resected specimens, when possible, or with the surgical findings. There were 42 gastric cancers and 44 primary gastric lymphomas. RESULTS: EUS made a correct diagnosis of cancer in 35 of 42 patients, with a sensitivity of 83%. Positive predictability was 87%, specificity was 97%, and negative predictability was 96%. Diagnostic accuracy was 95%. In the evaluation of cancer depth invasion, EUS was correct in 91% of cases. EUS displayed perigastric metastatic lymph nodes in 14 of 25 patients, with a sensitivity of 56%. Positive predictive value was 93%, specificity was 93%, and negative predictive value was 54%. Diagnostic accuracy was 69%. EUS made a correct diagnosis of lymphoma in 39 of 44 patients, with a sensitivity of 89%. Positive predictability was 87%, specificity was 97%, and negative predictability was 97%. Diagnostic accuracy was 95%. In the evaluation of lymphoma depth invasion, EUS was correct in 92% of cases. EUS displayed metastatic perigastric lymph nodes in 8 of 18 patients, with a sensitivity of 44%. Positive predictability was 100%, specificity was 100%, and negative predictability was 72%. Diagnostic accuracy was 77%. CONCLUSIONS: From these data it appears that in these diseases EUS has demonstrated specific ultrasonographic features that allow correct diagnosis and staging in the majority of patients. In difficult cases EUS may help to achieve the correct diagnosis. EUS also appear to be a useful tool for staging of gastric cancer and lymphoma. It shows not only tumor depth and local spread but also the passage from a pathologic to a normal wall and lymph node metastasis. With this accurate noninvasive staging procedure, in the near future many patients will no longer undergo exploratory laparotomy for surgical staging. Thanks to EUS, the choice of conservative or surgical treatment can be strongly affected. In case of surgery, EUS can orient the kind of surgical approach. Moreover, the use of EUS for evaluation of therapy during follow-up will probably become of major importance.     

Cystic meningiomas: unusual forms of intracranial neoplasms

OBJECTIVE: We report eight patients with cystic intracranial meningiomas to outline the neuroimaging spectrum of this unusual form of intracranial tumors. METHODS: Both CT and MRI were equally effective for the detection of peritumoral or intratumoral cystic lesions. However, neuroimaging findings were nonspecific and did not allow a correct preoperative diagnosis in most cases. CONCLUSIONS: Only a high index of suspicion permits the neurosurgeon a proper recognition of cystic meningiomas and its differentiation from the more common and malignant gliomas.     

Histology and cytology of laparoscopically operated "simple ovarian cysts"

In a retrospective study the histopathological findings of 127 laparoscopically operated unilocular anechoic smooth-walled ovarian cysts have been correlated with clinical characteristics (age, duration of observation, complaints, hormonal treatment), size by ultrasound, kind and colour of cysts content as well as cytological findings. The age of patients differed from 16-61 years (mean +/- s: 36 +/- 16). The histopathologic findings yielded 15 (11.8%) functional cysts, 30 (23.6%) persistent corpus luteum cysts, 9 (7.1%) endometriomas, 7 (5.5%) cystic teratomas, 9 (7.1%) undifferenciated cysts and 57 (44.9%) cystadenomas. There were no differences between histopathologic diagnosis groups according to age and cysts size by ultrasound. Functional cysts with complaints (n = 6) may explain that the observation time in 60% of all functional cysts was smaller than 6 weeks, whereas persistent corpus luteum cysts, endometriomas, cystic teratoma and cystadenomas had been observed for longer than 6 weeks in more than two thirds. Intraoperative evaluation of cysts content as "chocolate"-like was suspicious of endometriomas, but was also present in cysts of other histopathological findings. By means of cytology, endometrioma (siderophages) was suspected in 44.4% and a cystadenoma in 42.1% of all histopathologically verified cases. In all, the cytologic findings were useful for correct histopathological diagnosis in only 33.9% of all 127 cases. It is concluded that differential diagnosis of simple ovarian cysts is not possible by clinical characteristics, neither by ultrasound nor by cytological evaluation. Ovarian cysts should be observed for at least two hormonal cycles. A hormonal treatment by combination preparations containing high doses of oestrogen is also recommended. In cases of persisting ovarian cysts laparoscopic removal is necessary.     

Diagnosis of ectopic pregnancy--why we need a protocol

OBJECTIVE: To audit the management after instituting a screening programme for ectopic pregnancy in an institution with a protocol utilising ultrasound examination and serial human chorionic gonadotropin (hCG) and to examine the risk of missed diagnosis with deviation from the protocol. MATERIAL AND METHOD: A retrospective analysis of the management of 145 symptomatic patients in early pregnancies without intrauterine gestational sacs from ultrasound examinations, during the period April to June 1994 in Kandang Kerbau Hospital. Patients underwent serial hCG tests over 48 hours with or without repeat ultrasound scans before definitive treatment unless clinical indications for emergency surgery was necessary. RESULTS: There were 35 ectopic pregnancies (24%), 16 were viable intrauterine pregnancies (11%), 87 were non-viable pregnancies (60%) and 7 were of unknown outcome. There were much practice deviations from the protocol. Forty-four percent (64 cases) of the management decisions were made based on the initial clinical and ultrasound findings, and another 14% (21 cases) after a repeat assessment within the next day by either a repeat scan or serial serum hCG over one day. Among them, two of the 29 operated for suspected ectopic pregnancy were not ectopic (7%) and two of the 56 thought not to be ectopic, turned out to be ectopic (4%) (p < 10(-8)). Six percent (8 cases) defaulted after the initial assessments and one of them was found to be ectopic subsequently. Thirty percent (43 cases) adhered to the protocol. They had serial serum hCG done over two days. Seven of them requiring further repeats of serial serum hCG before management decisions were made. Four patients who were operated on were confirmed ectopic and 39 patients not operated on were not ectopic. Three percent (5 cases) were managed by serial hCG over 3 to 5 days and another 3% (4 cases) by repeating scan over one to two weeks without serial hCG. None of these was ectopic. The percentage change of hCG levels over two days gave indications of the likely diagnosis. CONCLUSION: Adhering to a protocol utilising the principle of ultrasound scan, serial hCGs and selective repeat ultrasound scans are highly recommended for the diagnosis of ectopic pregnancy. Any deviation from protocol is dangerous, with a 4% risk of missing an ectopic and a 7% risk of unnecessary operation for suspected ectopic pregnancy.