Radical surgery following inadequate drainage procedures for cystic neoplasms of the pancreas

Full Text

BACKGROUND

Pancreatic cystic lesions (PCL) represent the inhomogeneous group of diseases, characterized by the presence of a cavity with liquid content, located within the parenchyma of the pancreas or within the tissues adjacent to it. The distinctive feature of PCL comparing to the simple fluid accumulation foci is, generally, the presence of a capsule or a pseudocapsule (the wall) with various degree of intensity [1]. In general, PCL are detected approximately in 2–3% of the adult population, however, in the age groups of individuals older than 50–60 years, the incidence rate increases up to 10–15% and more [2, 3].

All the PCL can be divided into two main categories: the pseudocysts and the pancreatic cystic neoplasms (PCNs). Pseudocysts, being the most widespread type of PCL (80–90%), usually develop as a complication of acute or chronic pancreatitis or as a result of trauma of the pancreas [4]. The pathogenesis and the morphological features of pseudocysts determine the variable treatment type depending on the specific clinical situation [4–6]. In the settings of the wide implementation of endoscopic or other minimally invasive methods for draining purposes, the pancreatic pseudocysts in the majority clinical observations are subject to the effective sanitation-drainage procedures with the predominance of minimally invasive surgical approaches [7, 8].

The true PCL represent an average of 10–15% of all the PCL cases [2, 9], include the benign forms, such as the serous cystadenomas (SCA), and the neoplastic types, such as mucinous cystic neoplasms (MCN), the completely cystous neuroendocrine neoplasms of the pancreas (PanNEN) and the intraductal papillary mucinous neoplasms (IPMN), which have a higher risk of malignization, as well as the solid pseudopapillary neoplasms (SPN), which are classified as malignant neoplasms, but in the vast majority of cases they rarely spread the metastases and show the favorable prognosis in the settings of the timely and radical surgical treatment [10, 11]. Each of these categories of cystic neoplasms has its own risk of malignization level and requires a special clinical approach [12]. Thus, while the SCA are showing the extremely low probability of malignization (not more than 1% of the observations) [13], the MCN have the risk of malignization being 10–20% of the cases [14], also increasing with the presence of the solid component, of the inhomogeneity of the cystic neoplasm wall and of its enlargement [14, 15].

IPMN, representing 41–56% of all the cystic neoplasms in the pancreas, develop within the ductal system and differ by the potential risk of developing the invasive pancreatic carcinoma [10, 16, 17]. There are two main types of IPMN: the main-duct IPMN (MD-IPMN) and the branch-duct IPMN (BD-IPMN), with the MD-IPMN, generally, being more dangerous in terms of the risk of malignization [10, 17, 18]. The risk of malignization increases with the degree of the dilation of the main pancreatic duct and with the presence of solid growth inside it [16, 19]. The involvement of the main pancreatic duct is classified as the variant with high risk of malignization (approximately 40–70%) and the duct diameter of ≥10 mm, in accordance with the acting international recommendations, is considered as the indication for resection, while the duct dilation within the range of 5–9 mm is categorized as the worrisome features and requires a deeper evaluation [10, 16, 20]. The involvement of the lateral duct branches is characterized by the least risk of malignization (5–15%) [11, 18]. The mixed-type IPMN (MT-IPMN), which can involve both parts of the ductal system, have the medium risk of developing a carcinoma in situ (35–60%) [18, 19, 21].

The pancreatic cystic lesions can be quite easily diagnosed during the transabdominal ultrasound (US), however, in the majority of cases, for the purpose of defining the cyst type (pseudocysts or cystic neoplasms and its variants), one needs to use the more precise cross sectional imaging [22–24]. They include the computed tomography (CT) with multiphase contrast enhancement, the magnetic resonance imaging (MRI) with the mode of magnetic resonance cholangiopancreatography and the endoscopic ultrasound (USE) with the possibility of arranging the fine-needle aspiration for obtaining the cystic fluid sample for testing it for tumor markers (carcinoembryonic antigen, СЕА >192 ng/ml), as well as for the biochemical, the cytological and the molecular-genetic testing (KRAS/GNAS mutations) [25–27].

The treatment of PCL must depend on its specific type [28, 29]. In some types of PCL, the treatment tactics is undoubtful. Thus, in cases of pseudocysts, in the majority of cases the drainage procedures are indicated with selecting the optimal access and with taking into consideration the currently available possibilities of minimally invasive surgery (endoscopic, percutaneous under ultrasound guidance and laparoscopic) [8, 30, 31]. In cases of SPN, there are unambiguous indications for the resection surgery depending on the location and the size of tumor [12]. In other types of PCNs, the tactics of treatment or monitoring the lesion depends on multiple specific factors [22, 32, 33]. Currently, there are several national and international consensuses on the treatment of cystic pancreatic tumors [21, 28, 29], in which the algorithms were stipulated for selecting the tactics depending on the specific clinical situation. However, despite the standardized algorithms of examination, the differentiation between the pseudocysts and the PCNs at the pre-operative stage is not always successful, which is mainly related to the variability of their clinical-radiological signs and to the experience of a specialist [34]. As of today, in clinical practice we can still see the cases when the cystic pancreatic tumors at the pre-hospital or at the intraoperative stage are considered as pseudocysts, as a result of which, drainage procedures are done, which results in the delayed diagnostics of cystic neoplasm and the necessity for the repeated intervention, this time the resection one [35–37].

Research Aim — to present the experience of treating the PCL, which previously were diagnosed as “pancreatic cysts” or cysts in other locations, and treated by means of various non-radical draining surgical interventions.

METHODS

Research design

A retrospective observational single-center research was arranged with the analysis including the results of surgical treatment of PCL patients, which previously underwent various non-radical draining surgical interventions.

Conformity Criteria

Inclusion criteria: clinical-instrumental signs of PCL (MCN, IPMN, SPN, pancreatic adenocarcinoma); the age starting from 18 years and older; the presence of previous various draining surgical interventions due to having pancreatic cysts.

Non-inclusion criteria: pancreatic pseudocysts; retention cysts of the pancreas; signs of acute pancreatitis/exacerbation of chronic pancreatitis.

Research facilities

The research work was carried out within the premises of the Abdominal Surgery Department of the Federal State Budgetary Institution “National Medical Research Center of Surgery named after A.V. Vishnevsky” under the Ministry of Health of the Russian Federation (FSBI NMRCS named after A.V. Vishenvsky). The research included the patients that were admitted from January 2016 until February 2025.

Research description

The following parameters were evaluated:

• pre-operative period: the date of the first diagnosing the disease; the diagnosis that acted as an indication for draining surgery; the time from the date of diagnostics to the date of the first draining surgery; the type of conducted draining intervention and its result (duration of draining tube placement; decreasing/resolving of PCL; persisting draining tube with the discharge); date of the repeated diagnostics (recurrence) of PCL; diagnosis, with which the patient was undergoing the radical surgery; the time from the moment of diagnostics to the date of radical surgery;
• postoperative period: immediate results of surgical interventions, such as the total number and the type of postoperative complications, distributed acc. to the classification by Clavien–Dindo (2004) [38]; specific complications for pancreatic surgery, stratified using the classifications from the International Study Group for Pancreatic Surgery (ISGPS); length of postoperative hospital stay; definitive diagnosis according to histological examination data obtained for the resected samples [39–41].

Statistical analysis

The statistical processing was done using the IBM SPSS Statistics v28.0 (IBM Corp., Armonk, NY, USA). The normality of distribution for quantitative data was verified using the Shapiro–Wilk test. Taking into consideration the small sample size and the abnormal distribution, the quantitative parameters were presented as the median and the interquartile interval (Me [Q₁; Q₃]). The comparison of independent groups was carried out using the Mann–Whitney test; the categorical variables were compared using the Pearson’s χ² test or the Fisher’s exact test for the expected frequencies being < 5. For the evaluation of the correlations between the quantitative variables, the Spearman rank correlation was used (ρ); in case of normal distribution, the Pearson’s correlation was additionally calculated (r). The level of statistical significance was the double-sided p < 0.05. The overall accuracy of the pre-operative diagnosis was defined as the percentage of cases of complete matching for the pre-operative and the definitive morphological diagnosis; 95% confidence intervals (95% CI) for the percentages calculated using the exact binomial Clopper–Pearson method.

RESULTS

Research sample (participants)

From January 2016 until February 2025, at the Abdominal Surgery Department of the FSBI NMRCS named after A.V. Vishenvsky, due to having the PCL, a total of 177 patients were operated. The retrospective analysis involved the data from 19 (10.7%) patients, which previously underwent surgical interventions due to having the PCL. Among the 19 patients, 2 were men (10.5%), the number of women was 17 (89.5%). The mean age of the patients was 46.6 [27; 74] years: for men — 49.5 [38; 61], for women — 46.2 [27; 74].

Upon the revision of the anamnestic data, there was no information on the previous episode of acute pancreatitis or on the presence of chronic pancreatic disease until the moment of the first detection of PCL in any of the patients. Eight (42.1%) patients at the moment of detecting the PCL had clinical manifestations expressed as the epigastric pain syndrome, nausea, vomiting and dyspepsia. In the 11 (58%) patients, the symptoms were revealed occasionally during the regular medical checkups or during the examination due to the complaints not related to the disease of the pancreas.

The majority of patients (15; 79%) until the first draining surgical intervention had only the transabdominal ultrasound done, and the diagnosis was based only on a single visualization method. In 4 (21%) cases, the additional procedures included the CT and/or MRI, but in none of the cases the presence of cystic neoplasm of the pancreas was suspected. The dimensions of the PCL, according to the radiodiagnostics data, varied from 45 mm to 110 mm (84 [68.5; 99]).

The time from the moment of detecting the PCL to the first draining surgical intervention varied from 2 to 147 months (36 [23.5; 84]).

In 2 (10.5%) cases, there was a consecutive chain of two various types of draining surgical interventions due to the presence of the pancreatic cyst: in the first case, the initial procedure was the endoscopic cystogastrostomy, and then, due to the stent migration — the percutaneous external draining of the cyst under the ultrasound guidance; in the second case, there were two episodes of consecutive laparoscopic excision of the cyst walls. Thus, the total number of 19 patients had 21 surgical interventions, with this, in all the patients, the pre-operative diagnosis was defined as the “Pancreatic cyst” without its more detailed description. The spectrum of initially conducted PCL surgeries is provided in table 1.

 

Table 1 The characterization of the primary draining surgical interventions performed due to the presence of the pancreatic cyst

Surgical intervention	SCA	MCN	IPMN	SPN	PAC	Total
Endoscopic cystogastric anastomosis		1		1		2
Therapeutic puncturing under ultrasonography guidance				1		1
External draining under ultrasonography guidance		7	1	1	1	10
Laparoscopic excision of the cyst		2				2
Robot-assisted resection of the cyst in the pancreatic wall		1				1
Laparotomy, external draining of the cyst		1				1
Laparotomy, cystojejunostomy	1	3				4
Total	1	15	1	3	1	21

Note. SCA — serous cystadenoma; MCN — mucinous cystic neoplasm; IPMN — intraductal papillary mucinous neoplasm; SPN — solid pseudopapillary neoplasm; PAC — pancreatic adenocarcinoma.

 

Primary research findings

Endoscopic cystogastric anastomosis was done in 2 (10.5%) PCL patients. In one of the patients with a diameter of the cyst in the pancreatic tail being 5 cm, the plastic Pig Tail type stent was installed, which was removed endoscopically after 3 months due to the resolving of the cystic (fluidic) component based on the results of the control endo-ultrasound examination. The recurrence of the pancreatic tail cyst has occurred in 6 months after removing the stent with a background of a relapse of the pain syndrome in the left subcostal area and it was confirmed by the MRI findings. In the second case of the cystic lesion in the pancreatic head with the size of 4 cm, an endoscopic cystogastrostomy followed with the installation of the metallic self-expanding stent, which, in several days after the installation, has completely migrated into the gastric lumen and was removed, while the PCL was percutaneously drained under ultrasound guidance.

Single percutaneous puncture and drainage procedures under the ultrasound control were conducted in 11 (58%) patients: in 1 (5.3%) case — the therapeutic puncturing of the cystic lesion with the evacuation of fluidic fraction from the cyst lumen and with testing only amylase activity of the exudate; 10 (52.6%) cases were undergoing the percutaneous external draining of the cyst cavity. In this group of patients, the dimensions of the cystic lesions varied from 45 mm to 110 mm (84 [68.5; 99]). In 7 (36.8%) patients, the draining catheters were removed for the reason of cessation of discharge production and due to the absence of cyst cavity according to the data from the control ultrasound examination within the timelines from 2 weeks to 6 months (1 [0.625; 3.5] months); in 3 (15.8%) cases, the drainage tubes were left in place due to the presence of discharge and due to the presence of the residual cyst cavity according to data obtained by means of the ultrasound examination. In patients, in which the draining catheter was removed, the PCL recurrence has developed during the period ranging from 1 to 54 months (7 [4.5; 33]).

In one case (5.3%), the female patient was diagnosed with the cystic lesion with the diameter of 100 mm, located in the area of the pancreatic tail and in the retroperitoneal space, which, according to data obtained by means of ultrasound examination, due to its close proximity to the left kidney, was interpreted as the left kidney cyst. At the Urology In-Patient Department, the female patient underwent the laparoscopic excision of the cyst walls. In 6 months, upon the control ultrasound, a recurrence of the cystic lesion was diagnosed, due to which, a repeated laparoscopy was done with the excision of the cyst walls. The persisting pain syndrome has served as the justification for another ultrasound examination in 3 months, which once again has shown the cyst recurrence. According to the CT data, the diagnosis set was the cystic neoplasm of the pancreatic body and tail, more corresponding to the MCN.

In another patient (5.3%) with the pre-operative diagnosis of MCN of the pancreatic body with the size of 50×60 mm, during the robot-assisted intervention and during the intraoperative evaluation, a thin-walled cyst was found, deeply submerged into the pancreatic parenchyma and macroscopically looking like SCA. Due to the location and the syntopy of the lesion, its complete resection would require the subtotal corporocaudal resection of the pancreas. The decision was made on excising the freely located cyst wall and on further emergency histological examination, according to the data from which, the lesion was containing the fibrous tissue without the epithelial lining, due to which, the residual cavity was treated using the electrocoagulation, and the surgery was completed. However, in 13 months, the patient has developed a recurrence of the cystic lesion.

The past history of laparotomy was reported for 5 (26.3%) patients with the cystic lesions in the tail of pancreas, initially evaluated as pseudocysts. The dimensions of the lesions in this group of patients varied from 90 mm to 140 mm (110 [100; 120]). In one case (5.3%), the surgical intervention was the external draining of the cyst with further removal of the draining tube in 1.5 months, after which and another 2 weeks later, the clinical-instrumental recurrence was registered. Laparotomy with cystoenterostomy was performed in four (21%) patients. In 3 (15.8%) cases, the anastomosis was created on a Roux-en-Y jejunal loop, and in 1 (5.3%) case, on a Braun loop. In one patient (5.3%), despite the constructed anastomosis, a residual cystic cavity persisted according to ultrasound findings. Recurrence of the cystic lesion was observed in three (15.8%) patients during a follow-up period ranging from 13 to 72 months (median 3 months [1.75; 37.5]).

In general, 13 (68.4%) patients during the postoperative period had draining catheters in the cyst cavity, of which, in 10 (76.9%) cases they were removed within a period from 2 weeks to 6 months (1.25 [0.81; 3] months), in 3 (23.1%) cases, the drainage tubes remained in place until the moment of the radical surgical intervention due to the continued production of discharge.

The interval from the moment of the primary intervention to the clinically and instrumentally verified recurrence of PCL ranged from 0.5 to 84 months (6 [0.75; 13]).

In patients with minimally invasive interventions (puncture and draining, endoscopic), the median time to recurrence was 8 [3.5; 29] months, while after the laparotomic surgeries — 14 [5.5; 38] months (p=0.047). In patients with further removal of the draining catheter, the median time to recurrence was 6 [0.75; 13] months, and in the group without draining during the initial intervention — 10.5 [2; 25.75] months.

All the 19 patients, upon presenting to the FSBI NMRCS named after A.V. Vishenvsky, after an additional examination including the CT (in 17; 89.5%), the MRI (in 7; 36.8%), the CT combined with MRI (in 5; 26.3%), as well as additional endo-ultrasound examination (in 2; 10.5%) were diagnosed with the cystic neoplasm of the pancreas: MCN — 15 (78.9%), MD-IPMN — 2 (10.5%), SPN — 1 (5.3%), pancreatic ductal adenocarcinoma was suspected in 1 patient (5.3%).

All the patients subsequently underwent the radical surgery (open-access, laparoscopic or robot-assisted). In general, the time interval from the moment of the primary intervention due to the presence of pancreatic cyst and to conducting the resection surgery ranged from 2 to 144 months (14 [5.5; 38]). After the percutaneous puncturing and drainage procedures under ultrasound guidance, this interval ranged from 2 to 108 months (8 [3.5; 29]), after creating the co-junction using the endoscopic access — 6 months, after the laparoscopic excision of the cyst — 106 months, after the external draining of the cyst via the laparotomic access — 6 months (p=0.047). The types of conducted radical resection-type interventions are provided in table 2.

 

Table 2 Type of radical surgery

Type of surgery	SCA	MCN	IPMN	SPN	PAC	Total
Pancreaticoduodenectomy	1		1	1	1	4
Spleen-preserving distal pancreatectomy		1		1		2
Spleen-preserving distal pancreatectomy		5				5
Laparoscopic spleen-preserving distal pancreatectomy		3				3
Laparoscopic distal pancreatectomy with splenectomy		1				1
Robot-assisted spleen-preserving distal pancreatectomy		3				3
Robot-assisted distal pancreatectomy with splenectomy		1				1
Total	1	14	1	2	1	19

Note. SCA — serous cystadenoma; MCN — mucinous cystic neoplasm; IPMN — intraductal papillary mucinous neoplasm; SPN — solid pseudopapillary neoplasm; PAC — pancreatic adenocarcinoma.

 

Postoperative complications were registered in 9 (47.4%) cases, severe complications (according to the Clavien–Dindo classification ≥III) — in 5 (26.3%) patients. The distribution by the severity degree was the following: grade I complications — in 4 (21.1%) patients (including the suppuration of the post-operative wound and the external pancreatic fistula of B1 type), grade IIIa — in 2 (10.5%), grade IIIb — in 3 (15.8%). No deaths were reported during the postoperative period.

The specific complications, such as pancreatic fistula type B, were registered in 5 (26.3%) patients, of which there were 3 cases of type B1 and 2 cases of type B3 fistula. In all the observations, the fistulas have resolved spontaneously, the draining catheters were removed in the out-patient settings within the maximal period of 3 months after discharge. The postoperative intra-abdominal ISGPS class C hemorrhage has developed in 3 (15.8%) patients and required the repeated laparotomy with achieving the final hemostasis. In 2 cases, there was the combination of the pancreatic fistula and the arrosive intra-abdominal hemorrhage.

Length of postoperative hospital stay ranged from 6 to 30 days (14 [11; 20]): from 6 to 24 days (12 [9.25; 13.75]) for the uncomplicated and from 9 to 31 days (18 [15; 21]) for the complicated course of the postoperative period (p=0.003).

Based on histopathological examination, the diagnosis of MCN was established in 14 (73.7%) cases, SPN — in 2 (10.5%); one case (5.3%) each of SCA, MD-IPMN and ductal adenocarcinoma were identified.

The comparison of the pre-operation and the definitive diagnoses has shown that the diagnosis of MCN was confirmed in 13 (86.7%) cases out of the 15 preliminary diagnoses; in one case, the MCN was reviewed in favor of the SCA, in another one — in favor of the SPN. Out of 2 patients with the suggested MD-IPMN, in one case the diagnosis was confirmed morphologically, while in another case, based on the results of histological examination, MCN was diagnosed. The preoperative diagnoses of SPN (1 case) and of the ductal adenocarcinoma (1 case), according to the data from the cross sectional imaging, were matching the morphological conclusion. Thus, the total accuracy of preoperative setting the diagnosis using all the necessary cross sectional imaging, according to our data, was 84.2% (95% CI 62.4–94.5).

DISCUSSION

PCL represent a heterogeneous group of abnormalities, among which, the prevailing ones are the pseudocysts (80–90%), while the cystic neoplasms represent a total of 10–20% [1, 14, 36]. The latter may significantly vary in terms of their malignancy potential, which defines the necessity of differentiated diagnostic and therapeutic approaches [25, 33, 42]. Despite the significant progress in the field of diagnostic visualization and laboratory tests, the accurate diagnostics of the pancreatic cystic neoplasm remains a difficult task, which may lead, on the one hand, to excessive treatment, while on the other — to the untimely surgical intervention or its absence right when it is needed [20, 25, 43]. The errors in the differential diagnostics of PCL, resulting in the arrangement of the non-radical drainage procedures in cases of PCNs, remain a topical problem. In our series of 177 patients, operated due to having the cystic neoplasms in the pancreas, 19 cases (10.7%) previously underwent such surgical procedures, which corresponds to the rate of similar cases in the literature (5–15% at the specialized centers) [2, 36, 37]. If the cystic tumor of pancreas requires treatment, than, in the vast majority of cases, it is subject to complete removal using the resection surgeries [44–46], with this, in rare cases of SCA and MCN, there remains a place for such minimally invasive procedures as sclerotherapy under the ultrasonography guidance, during which, into the cystic cavity, a fine needle is introduced for fluid removal and for administering the sclerosing agents, promoting to the decrease of the cyst cavity [44, 47, 48]. This is applicable in cases of verified SCA and MCN of pancreas with their clinical manifestations and/or with the increased dimensions, in the settings of the technical accessibility for the puncture intervention, as well as in cases of high surgical risk or if the patient refuses to undergo the resection [49–51].

The predominance of women (88.2%) aged 27–74 years (with the median of 44 years) is typical for MCN, representing 73.6% of the cases. The variations of the dimensions of the cystic neoplasm were from 45 mm to 110 mm, with the absence of the past history of acute or chronic pancreatitis in all the patients emphasizes the absence of pre-conditions for setting the diagnosis of pseudocyst [8, 31, 45].

The primary interventions (endoscopic, percutaneous, laparotomic) were accompanied by 100% recurrence with the periods ranging from 2 weeks to 84 months (6 [0.75; 13]), which matches to the high recurrence capabilities of cystic neoplasms after drainage procedures claimed in the literature data (50–90% for MCN and IPMN) [35–37].

The repeated surgical procedures in all the patients included the pancreatic resection surgeries (distal pancreatectomy — 78.9%, pancreaticoduodenectomy — 15.8%), including the ones carried out using the minimally invasive access (laparoscopic — 52.6%, robot-assisted— 15.8%). The median time from primary to the repeated intervention was 14 [5.5; 38] months. The immediate results of the radical surgery are comparable to the literature data [3, 47, 52].

The distribution of the variants of pancreatic cystic neoplasms determined based on the results of morphological examinations is also comparable to the literature data [9, 25, 42]. Some tumors, such as pancreatic adenocarcinoma and the neuroendocrine tumors, can undergo cystic degeneration and still look like the cystic lesions [9, 14, 42]. The pre-operative structure of cystic lesions in our series was dislocated towards the MCN: at the stage of the additional examination at the tertiary surgical center, the diagnosis of pancreatic cystic neoplasm was set in all the patients, with this, the prevailing type was the MCN (15; 79%), significantly less frequent was the SPN (1; 5.3%), the MD-IPMN (2; 10.5%), as well as the pancreatic ductal adenocarcinoma (1; 5.3%). Such a type of preoperative findings generally correspond to the current data, according to which, at the specialized centers and upon the targeted examination, the prevailing ones are the MCN and the IPMN, while the SCA and rare variants of PCL, including the SPN, at the stage of the visualization are misdiagnosed or considered the less probable differential-diagnostic variants [23, 33, 44]. Based on the results of the pathomorphological examination, MCN was verified in 14 (73.6%) cases, i.e. slightly less frequently than it was suggested at the clinical stage (79%), with this, the percentage of SPN has increased from one (5.3%) pre-operative case to 2 (10.5%) according to the histological examination data, which indicates the complexity of the precise differential diagnostics of SPN. Besides, one case (5.3%) was registered for SCA, MD-IPMN and ductal adenocarcinoma (5.3%) each. The postoperative morphological verification has detailed the structure of the pancreatic cystic neoplasm, which has a fundamental value for further follow-up and prognosis. Determining the nature of PCL before the initiation of surgical treatment allows for avoiding the accidental draining of the pancreatic cystic tumor, misdiagnosed as pancreatic pseudocyst, also avoiding such negative consequences as the inevitable recurrence of the disease, and in the worst scenario — the progression of the neoplastic process in case of its malignant type [8, 25, 37].

Research limitations

Limitation related to the research sample. The present research represents a retrospective analysis of a small highly selective group of patients (19 out of 177 operated due to the presence of PCL), hospitalized to the single tertiary surgical center. The sample included only the patients in which, after the primary draining intervention, a cystic neoplasm recurrence was verified, after which followed the radical resection, while the patients with a similar treatment history, but operated in other hospitals or not referred to the specialized center, were not included into the analysis. This creates a risk of selection bias: the percentage of patients with erroneous draining of the cystic neoplasm of the pancreas (10.7% of all the resections) can be both decreased (unregistered cases outside the center) and increased (concentrating the most complex cases in the reference institution). The small research sample size results in the broad confidence intervals for evaluating the recurrence rates and the structure of complications, which is why the obtained results should be considered as characteristic predominantly for the patients referred to the specialized center after non-radical interventions, but not for the whole population of PCL patients. The generalisability of the conclusions in terms of the level of the whole target population is limited.

Limitations related to the comparability of groups. The research was not suggesting the randomization or the preliminary planning of the full-scale comparable groups. The differences between the subgroups of patients (by the type of primary access — endoscopic, percutaneous, laparotomic; by the fact of removing or leaving the draining tube etc.) were formed as a result of clinical decisions made in various institutions and at various stages of treatment. Within this context, the baseline characteristics (the size and the location of the cystic tumor; the supposed diagnosis; the concomitant disease; the experience of the surgeon) could significantly differ, while the multiple intervening factors were not taken into account during the analysis. The demonstrated differences in the time recurrence between the groups (for example, after the minimally invasive and the laparotomic interventions) reflect the cumulative effect of these unaccounted factors and could not be interpreted as the cause-and-effect relation between the type of primary intervention and the risk of recurrence. Thus, the reference conditions for the comparability of the groups (д)–(е) for the observational comparative research works were not fully observed, which increases the risk of bias in the interpretation of the intergroup differences.

Limitations related to the research parameters. As the outcomes, the parameter used were clinically significant, but not unambiguously reconstructible: the fact and the timing of the cystic lesion recurrence; the time interval to the radical surgery; the rates and the structure of postoperative complications; the definitive histological diagnoses. The part of the key information on the primary intervention (the indications, the extent, the technical details, the dynamic changes of the cystic lesion after draining) was obtained retrospectively — according to the data from medical documentation from other institutions, which creates a risk of information bias and the incompleteness of data. The type of cystic lesion at the moment of the primary intervention in the majority of cases was not morphologically verified and was reconstructed post factum based on the results of the definitive pathomorphological examination of the resected tissue samples, which admits the theoretical possibility of non-conformity between the baseline data and the data determined later (for example, the evolution or the development of new tumor with a background of pre-existing lesion). This can result in the re-evaluation of the association between the draining intervention and further recurrence of the same tumor lesion with the same location. Besides, the analysis does not include the remote oncological outcomes (survival, the rate of remote recurrences), which is why the effect of delayed radical intervention on the long-term prognosis could be underestimated.

Limitations related to the measurements methods. The methods of visualization and laboratory diagnostics used before the primary intervention, were inhomogeneous: in the majority of patients, the diagnosis of pancreatic cyst was set only according to data obtained after the ultrasound examination, while the CT, the MRI and the endoscopic ultrasound examination were carried out not necessarily and at various stages. The quality of the primary examinations and the description protocols, performed at various treatment institutions, is impossible to standardize retrospectively, which limits the reproducibility of evaluating the baseline characteristics of PCNs and the degree of their difference comparing to the pseudocysts. Within a long-term inclusion period (2016–2025), there occurred the changes in the accessibility and the quality of cross sectional imaging, as well as in the surgical technique (the growth of the percentage of laparoscopic and robot-assisted resections), which could affect the rate of detecting the cystic neoplasms, the rates and the structure of complications, the duration of hospitalization. Due to the small sample size, the effects of the time factor and the learning curve were not analyzed separately, which could bias the evaluation of the immediate results towards the better values for the later observations.

At the same time, for the evaluation of the postoperative complications, the standardized and the validated scales (Clavien–Dindo, ISGPS classification) were used, which increases the comparability of our data to the results from other centers, however, the other spectrum of complications (mild events treated in the out-patient settings) could be underestimated, for their retrospective registration is less sensitive than the prospective monitoring.

On an aggregate basis, the listed limitations introduce the high degree of skepticism in terms of the absolute values of the main parameters (rate of recurrence after the drainage procedures; the percentage of patients with erroneously conducted draining; the structure of complications). The most probable directions of bias include the overstating of the evaluation of recurrences and of the complexity of further resections for the whole population of patients with drained PCL and, on the contrary, the decrease of the true rate of such cases in the general cohort of patients with cystous neoplasms, taking into consideration the selective type of the sample. Due to that, the obtained results should be interpreted predominantly as the characteristics of the “marginal” clinical scenarios observed in the highly specialized center, while the conclusions on the risks of erroneous tactics in cases of PCNs must be compared to the data from other retrospective series and recommendations, based on larger samples.

CONCLUSION

Upon detecting the cystic lesions in the pancreas, it is necessary to employ all the available diagnostic possibilities in order to set the correct diagnosis and to primarily arrange the differential diagnostics between the pseudocyst and the cystic tumor. The erroneously performed drainage surgical procedures in cases of PCNs do not result in the recovery and in the vast majority of cases they are accompanied by the recurrence of the disease, and in a number of cases they can critically affect the timelines of initiating the radical therapy, which is directly related to the long-term prognosis; they also complicate the further conduct of radical resection procedures.

Thus, the timely setting the correct diagnosis of cystic tumor and the radical surgical treatment provide the favorable outcome and further prognosis.

ADDITIONAL INFORMATION

Author contributions: P.V. Markov, defining the concept and the research design, editing the article text, responsibility for the integrity of all article parts; А.V. Mamoshin, collecting and processing the material, analysis of literature sources, writing the article text, preparing the article for publication; V.Yu. Struchkov, validation, reviewing and editing the manuscript; О.R. Arutyunov, collecting and processing the material; М.V. Dvukhzhilov, collecting and statistical processing of the material; А.I. Burmistrov, collecting and processing of the material. Thereby, all authors provided approval of the version to be published and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Ethics approval: The research protocol was approved by the local Ethics Committee of the Federal State Budgetary Institution “National Medical Research Center of Surgery named after A.V. Vishenvsky” (protocol No. 010-2025, dated 24.10.2025). All the research participants have voluntarily signed the informed consent form prior to the inclusion into the research program.

Funding source: The research work was carried out within the framework of the scientific-research work by the State Assignment of the Ministry of Health of the Russian Federation: R&D No. 123020700067-2 “New technologies and personalized approach in the combined differential diagnostics and surgical treatment of focal lesions in the pancreas”, conducted within the premises of the Federal State Budgetary Institution “National Medical Research Center of Surgery named after A.V. Vishenvsky” (Russian Federation). Type of funding: financial support of the R&D activities, provision of the research infrastructure of the center (equipment, expendables, software). No additional targeted financing, including the preparation of the publication, was provided to the authors. The organization acting as a source of financing, did not participate in compiling the research design, in collecting, analyzing or interpreting the data, in preparing the manuscript and in making the decision on submitting the article to the journal; no limitations were implemented regarding the usage of data and the distribution of the research results from the source of funding.

Disclosure of interests: The authors declare the absence of clear or potential conflicts of interests related to the publication of the present article.

Statement of originality: The present article is an original research work. It contains the research results, including the clinical material that have not been previously published or used in other scientific articles, theses or reports. The clinical data were initially collected within the framework of the routine medical aid and further follow-up of the patients at the FSBI “National Medical Research Center of Surgery named after A.V. Vishenvsky”, while their systematization, statistical processing and interpretation conducted for the first time for the purpose of the present research. The article text, the tables and figures are original and do not reproduce any previously published materials from the authors or other persons completely or partially. When using the data of the previously published sources (data, terms, figures), they are provided solely in the summarized for in the literature review section with proper attribution and full bibliographic citation of the primary sources, in strict conformity with the conditions of use established by the rights holders. This research work does not reuse the authors’ own previously published data, text, tables, or figures, nor the published material from the third party in the form of direct reproduction. The information used in the article is not subject to any restrictions on its use or distribution by the third parties.

Data availability statement: The authors state that all aggregated data necessary to understand and reproduce the main results of the research are presented in the article and/or in its tables and figures. The individual-level clinical data (impersonalized patient records, examination results, and procedural details) obtained during the course of this research cannot be made publicly available due to the data protection legislation of the Russian Federation and due to the internal regulations of the institution. The authors are willing to provide a limited access to the impersonalized subsets of the original dataset upon reasonable written request from researchers interested in verifying the reported findings or performing the joint analyses. The requests should be addressed to the corresponding author using the contact details provided in the manuscript. Data sharing is possible subject to approval by the local Ethics Committee of the FSBI “National Medical Research Center of Surgery named after A.V. Vishenvsky” under the Ministry of Health of the Russian Federation, and upon signing the appropriate non-disclosure and data use agreement. No fixed embargo period is imposed; each request will be addressed on a case-by-case basis.

Generative AI: Generative AI technologies were not used for this article creation.

About the authors

Pavel V. Markov

A.V. Vishnevsky National Medical Research Center of Surgery; Russian Medical Academy of Continuous Professional Education

Email: pvmarkov@mail.ru
ORCID iD: 0000-0002-9074-5676
SPIN-code: 6808-9492

MD, PhD

Russian Federation, Moscow; Moscow

Andrian V. Mamoshin

A.V. Vishnevsky National Medical Research Center of Surgery; Orel State University

Author for correspondence.
Email: dr.mamoshin@mail.ru
ORCID iD: 0000-0003-1787-5156
SPIN-code: 2553-1200

MD, PhD, Assistant Professor

Russian Federation, Moscow; Orel

Vladimir Yu. Struchkov

A.V. Vishnevsky National Medical Research Center of Surgery

Email: doc.struchkov@gmail.com
ORCID iD: 0000-0003-1555-1596
SPIN-code: 4996-7802

MD, PhD

Russian Federation, Moscow

Ovanes R. Arutyunov

A.V. Vishnevsky National Medical Research Center of Surgery

Email: arutyunov_ovanes@mail.ru
ORCID iD: 0000-0001-9425-1924
SPIN-code: 3553-7425

MD

Russian Federation, Moscow

Mikhail V. Dvukhzhilov

A.V. Vishnevsky National Medical Research Center of Surgery

Email: dr.dvukhzhilov@mail.ru
ORCID iD: 0000-0001-7283-7465
SPIN-code: 2025-4436

MD

Russian Federation, Moscow

AleksandrAleksandr I. Burmistrov

A.V. Vishnevsky National Medical Research Center of Surgery

Email: aibur3619@gmail.com
ORCID iD: 0000-0001-8853-3394
SPIN-code: 2674-0484

MD

Russian Federation, Moscow

References

Supplementary files