Determination of the diagnostic potential of research methods in the differentiation of pancreatic cancer and chronic pancreatitis based on evidence-based medicine

.

У даний час використання принципів доказової медицини є невід'ємною частиною оцінки ефективності та безпеки сучасної технології діагностики та лікування будь-якого захворювання чи клінічного Currently, the use of the principles of evidence-based medicine is an integral part of assessing the effectiveness and safety of modern technology for the diagnosis and treatment of any disease or clinical condition.Pancreatic cancer is no exception to this rule.When making a decision on the use of diagnostic methods for a patient with pancreatic cancer, doctors are guided not only by their own clinical experience but also, to a lesser extent, by the results of clinical studies, meta-analyses, randomized controlled trials or observational studies [1][2][3].
According to the literature data, there is an increasing trend in the development of pancreatic cancer, particularly intraductal adenocarcinoma, which is the fourth leading cause of cancer death among men and women in the United States of America, with an overall median 5-year survival rate of 8% and a median survival rate of 32% in localized disease [4,5].
The differential diagnosis of pancreatic cancer and chronic pancreatitis (CP) with a predominant lesion of the pancreatic head is one of the most controversial issues in surgical pancreatology [6][7][8].It is equally important to avoid delays in diagnosis that lead to the progression of undiagnosed cases of pancreatic cancer which can quickly become inoperable and lead to death.
According to the recommendations of the guidelines for the diagnosis and treatment of CP of the United European Gastroenterological Working Group [9], endoscopic ultrasound (EUS), magnetic resonance imaging (MRI), and computed tomography (CT) are the best imaging methods for establishing the diagnosis of CP.Contrast-enhanced CT is currently one of the most effective methods for diagnosing pancreatic cancer, with a sensitivity and specificity of up to 90% and 99%, respectively [10].Findings that may predict pancreatic cancer include (from lowest to highest specificity): ductal dilatation (sensitivity of 50% and specificity of 78%), hypoattenuation (sensitivity of 75% and specificity of 84%), interruption of the main pancreatic duct (sensitivity of 45% and specificity of 82%), distal pancreatic atrophy (sensitivity of 45% and specificity of 96%), abnormalities of the pancreatic contour (sensitivity of 15% and specificity of 92%) and dilation of the common bile duct (sensitivity of 5% and specificity of 92%) [11].MRI allows a tumor to be detected at the early stages, providing a comprehensive analysis of morphological changes in the parenchyma of the pancreas, as well as the main pancreatic duct.Alabousi M. et al. conducted 14 studies that included 987 patients with pancreatic cancer (205 with liver metastases) and found that the sensitivity for CT and MRI was 45% (confidence intervals [95% CI] 21-71%) and 83% (95% CI 74-88%), respectively.Specificity for CT and MRI was 94% (95% CI 84-98%) and 96% (95% CI 93-97%), respectively [12].
A meta-analysis of the effectiveness of pancreatic cancer diagnosis was performed by evaluating pancreatogram images using endoscopic retrograde pancreatography (ERCP).Its overall sensitivity and specificity were 57.9% and 90.6%, respectively [16].In the study of Ishii Y. et al., the sensitivity and specificity indicators of endoscopic ultrasonography and ERCP were compared.The combined sensitivity indicators of EUS, ERCP and EUS+ERCP were 76.7%, 57.9% and 79.9%, respectively; combined specificity indicators were 91.7%, 90.6%, and 94.2%, respectively [17].
Preoperative pathomorphological diagnosis is necessary to determine the diagnosis and subsequent treatment strategy for patients with pancreatic diseases.EUS-guided fine-needle aspiration biopsy is a first-line method of diagnosis, with a combined sensitivity and specificity of the method of 85-92% and 96-98%, respectively [18].
Thus, the differential diagnosis of pancreatic cancer and CP with predominant lesion of the pancreatic head is an urgent problem of modern pancreatology.The complexity of the issue lies in the fact that pancreatic cancer is often detected in patients with CP, but CP is not considered by scientists as a precancerous disease [20][21][22][23][24].The reduced number of cases of resectable pancreatic cancer indicates the need to develop programs for the differentiation of these pathologies with the wide implementation of modern diagnostic methods.
The objective of the work is to compare the results of diagnostic methods (which are standardly used in accordance with clinical guidelines for the diagnosis of chronic pancreatitis) based on evidence-based medicine and quantitative assessments of their specificity, sensitivity, and general accuracy in terms of the differential diagnosis of chronic pancreatitis and pancreatic cancer in order to determine the most significant results for the differentiation of these pathologies. of January 5, 2023) and the Municipal Health Care Institution of the Kharkiv Regional Council «Regional Clinical Hospital» for the period from 2015 to 2023.
The patients were examined in accordance with the guidelines developed by the multidisciplinary working group on the topic «Chronic pancreatitis» (approved by the order of the Ministry of Health of Ukraine of 18.12.2020No. 1908 (as amended by the order of the Ministry of Health of Ukraine of 25.01.2023No. 145).The patients underwent general clinical blood and urine tests, biochemical blood tests, and the following procedures and examinations available in the clinic: ultrasound examination of the abdominal organs, spiral computed tomography (CT), fibrogastroduodenoscopy, endoscopic retrograde cholangiopancreatography (ERCP), magnetic resonance imaging (MRI), CA 19-9 tumor marker test, ultrasound-guided percutaneous puncture of the pancreas, aspiration of pancreatic juice, urgent intraoperative biopsy, and histological examination of surgically resected areas of the pancreas.Based on the received data, we performed a comparison of their quality, accuracy and informativeness .
At the first stage, we studied patients' complaints and disease history, laboratory test data, and ultrasound of the pancreas and adjacent organs.Percutaneous puncture of the pancreas was performed in 16 patients, aspiration of pancreatic juice -in 8 patients.77 (96.3%) patients underwent spiral CT, 45 (56.3%) -MRI.The use of spiral CT and MRI made it possible to assess the condition of the parenchyma of the pancreas, the main pancreatic duct, parapancreatic tissue, and adjacent organs.At the second diagnostic stage, ERCP was used in 58 (72.5%) patients, contrast-enhanced X-ray examination of the alimentary canal was performed in 15 (18.8%) patients, and in 42 (52.5%)patients CA 19-9 tumor marker test was performed.Based on the received data, a preliminary diagnosis was established and a further treatment strategy was planned.
Cancer was suspected during ultrasound of the pancreas when there were space-occupying formations in the parenchyma of the gland or areas of altered pancreatic tissue.The main sonographic criteria for pancreatic cancer were a combination of such signs as the presence of a focal formation, reduced echogenicity, uneven contours and unclear boundaries, and heterogeneity of the parenchymal structure.Indirect sonographic signs of pancreatic cancer included the presence of pancreatic ectasia, enlarged para-aortic lymph nodes in the region of the hepatoduodenal ligament, and liver metastases.
Patients with obstruction and dilatation of the main pancreatic duct of the pancreas were subject to further examination with CT and ERCP.CT was used in 77 (96.3%) patients.The examination was used to assess the shape, size, structure of the gland, its relationship with the surrounding organs and tissues, the condition of the extrahepatic bile ducts, duodenum, parapancreatic tissue, and regional lymphatic vessels, and also for the differential diagnosis of tumor and non-tumor changes in the gland.We also estimated tissue density, contour, localization, condition of the main pancreatic duct and choledochus, the presence of cystic formations in the parenchyma of the pancreas, the condition of the duodenum and adjacent organs, which was vitally important for choosing surgical intervention tactics.Дослідження проведено відповідно до етичних стандартів Гельсінської декларації Всесвітньої медичної асоціації про етичні принципи проведення наукових медичних досліджень за участю людини (1964-2008 рр.), директиви Європейського товарист-MRI was used in 45 (56.3%) patients to assess the condition of the pancreas, adjacent organs, parapancreatic tissue, extraperitoneal space, and regional lymph nodes.
ERCP was performed in 58 (72.5%) patients in order to verify the nature of biliary hypertension and impaired patency of the main pancreatic duct.In the presence of mechanical jaundice and distal block in patients with pancreatic cancer and CP with a predominant lesion of the pancreatic head, transpapillary stenting with plastic stents with a diameter of 2.5-4.0 mm was considered appropriate.During primary drainage, a stent with a diameter of 8-9 Fr was mainly installed, followed by a papillotomy and replacement of the stent with a stent with a larger diameter of 10-12 Fr after the regression of jaundice.
Ultrasonography-or CT-guided percutaneous fineneedle aspiration biopsy of the pancreas was used in 16 (20%) patients, with aspiration of pancreatic juice in 8 (10%) patients, with the aim of differential diagnosis between pancreatic cancer and CP.Contraindications for the procedure were: mechanical jaundice -bilirubinemia over 50 μmol/l; increase in blood or urine amylase; the prothrombin index level is less than 50%; impaired evacuator function of the stomach, the phenomenon of stagnation in the upper areas of the gastrointestinal tract; a relative contraindication was the size of the lesions of the pancreas less than 3 cm.
Based on the received data, their quality, accuracy and informativeness were compared.When describing the data, absolute and relative numbers of cases were given (n (%)); 95% confidence intervals (CI95%) were also estimated for the relative frequencies of occurrence of clinical and diagnostic signs according to the Wilson method.Proportions of patients with a certain sign in chronic pancreatitis and pancreatic cancer were compared using the Z-test for two proportions at a confidence level of 0.95.Epitools service (Ausvet Ltd., Australia) was used for calculations [Sergeant, ESG, 2018.Epitools Epidemiological Calculators.Ausvet.Available at: http://epitools.ausvet.com.au].
The quality/accuracy/informativeness of the methods used in the research in the differential diagnosis of pancreatic cancer and CP were compared according to such parameters as overall accuracy, specificity, sensitivity, positive and negative predictive value (PPV, NPV), diagnostic odds ratio (DOR), Youden's index (J) and Matthews correlation coefficient value (ϕ).When evaluating these parameters, cases of pancreatic cancer were considered «positive» results, and cases of CP were considered «negative».To calculate the main of the listed characteristics and their confidence intervals, we used the web interface of the corresponding tool of the  [14], а КТ з контрастуванням -до 90% та 99% відповідно [10].
During the study, a comparative analysis of direct and indirect ultrasonographic and spiral CT features of pancreatic cancer and CP with predominant lesion of the pancreatic head was performed.It was established that the significant difference between these two pathologies was the increased or mixed echogenicity of the tissue, the preserved normal dimensions of the organ with clearness and unevenness of its boundaries.The main criteria for pancreatic cancer were such indicators as focal lesions, reduced echogenicity with uneven boundaries, and heterogeneity of the structure.Significant indirect signs of pancreatic cancer were: pancreatic ectasia, the presence of enlarged para-aortic lymph nodes and lymph nodes in the region of the hepatoduodenal ligament, liver metastases.
The qualitative indicators of the differential diagnosis of pancreatic cancer and CP for the analyzed examination methods are shown in Table 1.
During the study, it was established that the overall accuracy of ultrasound for the diagnosis of pancreatic cancer was 82.5% (the interval estimate of the overall accuracy was (72.38; 90.09)%), with a sensitivity of 76.32% (the sensitivity of the method was (59.76; 88.56)%)); and specificity of 88.10% (interval specificity -(74.37;96.02)%).The overall accuracy of spiral CT in the diagnosis of pancreatic cancer was 92.21% (interval estimate of overall accuracy was (83.81; 97.09)%, with sensitivity of 89.19% (interval sensitivity of the method -(74.58;96.97)%); and specificity of 95% (interval specificity -(83.08;99.39)%).In general, it should be noted that the obtained data did not differ significantly from the data of international multi-centre studies.According to the literature data, the sensitivity and specificity of ultrasound in the diagnosis of pancreatic cancer varies from 75% to 89% and 90% to 99%, respectively [14], and contrast-enhanced CT -with up to 90% and 99%, respectively [10].

Оригінальні дослідження
Original research respiratory section of the choledochus in CP and pancreatic cancer in our sample of patients was statistically insignificant (Z = 1.4,p = 0.0746 > 0.05).Unsuccessful contrast of the main pancreatic duct during ERCP occurred in 5 patients (6.3% with CI95% from 2.7% to 13.8%), which was associated with the block of its distal part and deformation of the duodenum.Thus, morphological signs of pancreatic cancer, namely block of the distal part of the main pancreatic duct, its stenosis, narrowing, and expansion of side branches, were also present in the pancreatograms of patients with this pathology.When analyzing pancreatograms of patients with pancreatic cancer, it was established that they differed from pancreatograms of patients with CP in the presence of strictures longer than 1 cm, the severity of changes in the main pancreatic duct and its side branches, and more extensive stenosis of the distal part of the common bile duct.The presence of calculus in the ducts of the pancreas and calcifications in its parenchyma did not affect the diagnosis of cancer.The overall accuracy of ERCP in cancer diagnosis was 71%, with sensitivity and specificity of the method of 78.6% and 63.3%, respectively.The sensitivity values of ERCP obtained by us were higher, and the specificity values were lower in comparison with the data presented in the meta-analysis of Li H. et al. [16], where the sensitivity and specificity of this method were 57.9% and 90.6%, respectively.The interval estimate of the accuracy of the method of transcutaneous biopsy of the pancreas for determining CP was (34.9; 96.8)%; for pancreatic cancer -(8.5; 75.5)%.The interval estimate of the accuracy of the method of endoscopic aspiration of pancreatic juice for CP was (6.8; 93.2)%; for pancreatic cancer - (19.4; 99.4)%.The overall accuracy of the method of transcutaneous biopsy of the pancreas in the diagnosis of cancer was 56.3%, with sensitivity and specificity of 37.5% and 75%, respectively.The overall accuracy of endoscopic aspiration of pancreatic juice was 62.5%; with sensitivity and specificity of 75% and 50%, respectively.Among all the used methods of differential diagnosis of pancreatic cancer and CP, only for these two methods the confidence intervals of the diagnostic odds ratio contained 1 (Table 1), and in the logarithmic scale -0 (Fig. 1), which indicates the lack of confirmation of the diagnostic value of these methods by the data of the studied sample of patients, although it may be due to its small size.
A comparative analysis of the qualitative parameters of the instrumental research methods used for the differential diagnosis of pancreatic cancer and CP (Table 1) shows that the spiral CT method has the best characteristics in this sense, which is shown by the highest value of the Youden's index and the highest Matthews correlation that indicate the best ratio of sensitivity and specificity and high conformity of diagnostic results with real data.Also, the highest diagnostic odds ratio was obtained for the spiral CT method in our sample of patients (Table 1).As an additional method of diagnosis, the measurement of the level of the CA 19-9 tumor marker over time was considered.An increase in the level of CA 19-9 was found in 40.5% of patients (CI 95% from 27.0% to 55.5%), while the average level was significantly higher in pancreatic cancer.During the study, it was found that the level of the CA 19-9 tumor marker depended on the size of the tumor and єнтів виявилася статистично незначущою (Z = 1,4, p = 0,0746 > 0,05).Невдалі випадки контрастування головної панкреатичної протоки при виконанні ЕРХПГ мали місце у 5 хворих (6,3% з ДІ95% від 2,7% до 13,8%), що було пов'язано з блоком її дистального відділу та деформацією дванадцятипалої кишки.In general, the use of a combination of instrumental diagnostic methods together with the measurement of the CA 19-9 tumor marker showed a higher quality of differential diagnosis of pancreatic cancer from CP compared to individual instrumental methods (Table 1, Fig. 1).
The role of urgent intraoperative biopsy for the differentiation of pancreatic cancer from CP is unquestionable, its use contributes to adequate surgical treatment, which is the key to long-term recurrence-free survival.Currently, there is a general tendency to reduce the volume of resected tissues in compliance with the principles of oncological safety, which improves the quality of life of patients and does not worsen the prognosis of the disease, therefore the number of intraoperative morphological studies is increasing.The main disadvantages of the study of frozen sections were the limited number of studied areas and the lower quality of histological preparations than in the planned histological examination after the preparation of paraffin blocks, which affected the accuracy of the morphological Роль термінової інтраопераційної біопсії для диференціації раку ПЗ та ХП безсумнівна, її використання сприяє адекватному проведенню хірургічного лікування, яке є запорукою тривалого безрецидивного виживання.У даний час відзначається загальна тенденція до зменшення обсягу резекційованих тканин із дотриманням принципів онкологічної безпеки, що покращує якість життя пацієнтів та не погіршує прогноз захворювання, тому кількість інтраопераційних морфологічних досліджень зростає.Основними недоліками дослідження заморожених зрізів були обмежена кількість досліджуваних ділянок та нижча якість гістологічних препаратів, ніж при плановому гістологічному дослідженні після приготування парафінових блоків, що впливало на точність
On the basis of evidence-based medicine, it was established that spiral computed tomography is the most informative method for the differential diagnosis of pancreatic cancer with chronic pancreatitis, with an overall accuracy of 92.2%, sensitivity of 89.2%, and specificity of 95%.The low qualitative indicators of ultrasound diagnostics can be increased due to its complex combination with the CA 19-9 tumor marker test with an overall accuracy of 92.3%, with sensitivity and specificity of 95% and 91%, respectively, which is significantly different from such indicators when the method was used separately (82.5%, 76.3%, and 88.1%, respectively).In the cases when the complex combination of endoscopic retrograde cholangiopancreatography with the measurement of the level of the CA 19-9 tumor marker was used, the overall accuracy of the method was 95.2%, with sensitivity of 95% and specificity of 95.5%, which is significantly different from such indicators obtained when the method was used separately (71%, 78.6% and 63.3%, respectively).Advances in modern imaging methods cannot replace histological examination as the «gold standard» for the correct determination of the morphological substrate of these pancreatic lesions.