Local resection for solid pseudopapillary neoplasms of the pancreas shows improved postoperative gastrointestinal function and reduced mental stress: a multiquestionnaire survey from a large cohort

Background: Solid pseudopapillary neoplasm (SPN) of the pancreas is a rare, low-grade malignant pancreatic tumor with a highly favorable prognosis. Most SPN patients are young and middle-aged women. The main controversial topic for SPN is local resection (LR) versus radical resection (RR). Theoretically, LR could lead to better gastrointestinal function (GIF) and less mental stress. However, no data is available to support this hypothesis. Methods: All SPN patients undergoing surgical treatment in Peking Union Medical College Hospital from 2001 to 2021 were included in the study. A cross-sectional online multiquestionnaire survey containing 110 questions was sent to them (Clinicaltrial.org, NCT05604716). This online multiquestionnaire survey focused on GIF and mental stress and consisted of eight questionnaires. Multiple linear regression analysis was conducted to identify independent factors impacting GIF and mental stress. Results: A total of 183 cases provided valid results. Among them, 46 patients (25.1%) underwent LR, and 137 (74.9%) underwent RR. Ninety-four cases (51.4%) underwent minimally invasive surgery (MIS), while 89 (48.6%) underwent open surgery. The average GSRS score of the patients was 1.9±0.7, indicating that most suffered from mild gastrointestinal dysfunction. The scores of PHQ-9 and GAD-7 in 16 patients (8.7%) and 27 (14.8%) patients, respectively, were beyond 10.0, which indicated clinical depression and anxiety. Additionally, 19 (10.4%) patients reported poor ability to work, and 31(16.9%) patients had significant body image concerns. Compared to other clinicopathological characteristics, LR (LR vs. RR: PHQ-9 score, P=0.018; WAI average score, P=0.010; EORTC QLQ-C30, nine subdomains, P<0.05; GSRS average score, P=0.006) and MIS (MIS vs. open surgery: EORTC QLQ-C30, three subdomains, P<0.05; GSRS average score, P=0.006) were the most significant factors predicting improved GIF and reduced mental stress. Conclusions: This study systematically presents postoperative GIF and mental stress of SPN patients using validated multiquestionnaires for the first time. It provides solid evidence that LR and MIS can improve GIF and reduce mental stress after surgery for SPN patients, which could be helpful for the surgeons to make more personalized surgical plans for their patients.


Introduction
Solid pseudopapillary neoplasm (SPN) of the pancreas is a rare pancreatic tumor comprising nearly 1-3% of all pancreatic tumors [1,2] .The incidence or detection rate of SPN has been on the rise due to the more widespread use of abdominal imaging [3][4][5][6] .The primary controversial topic surrounding SPN is the choice between local resection (LR) and radical resection (RR).In theory, LR is less invasive and could significantly preserve organ function, leading to better postoperative gastrointestinal function (GIF) and less mental stress.On the other hand, RR with lymph node dissection could achieve better oncological outcomes.Referring to the oncological outcome of SPN, numerous reports from high-volume centers have comprehensively demonstrated that the incidence of lymph node metastasis is extremely low.Consequently, LR and RR have comparable long-term oncological outcomes [3,[7][8][9][10] .However, data on postoperative GIF and mental stress of SPN patients have not been reported yet, and there is no supportive evidence regarding the impact of LR and RR on these aspects.Herein, the authors adopted validated questionnaires previously used in studies of other tumors after surgery [11][12][13] to assess GIF and mental stress in a large cohort of SPN patients using a novel online methodology of eight questionnaires.For the first time, this study provides information on SPN patients' GIF and mental stress using validated questionnaires.The results of this study can assist surgeons and patients in making better, individualized clinical decisions to improve GIF and reduce mental stress after surgery.

Patients and follow-up
This cross-sectional single-center study collected data from hospitalized SPN patients between October 2001 and April 2021 from the Peking Union Medical College Hospital electronic medical record system.The contact, demographic, and clinical-pathological information, including telephone number, age, sex, occupation, marital status, educational level, symptoms, diagnosis, surgical procedures and methods, postoperative complications, and pathological outcomes, were retrieved.All patients had a definitive pathological diagnosis of SPN.The definition and severity of complications were defined according to the International Study Group of Pancreatic Surgery (ISGPS) and the Clavien-Dindo classification [14,15] .The histopathological characteristics of the tumors included lymph node metastasis, distant metastasis, and organ and tissue invasion (fatty tissue, nerves, vessels, bile duct, duodenum, stomach, transverse colon, and others).
All patients with available telephone numbers were called and asked whether they could accept our online multiquestionnaire survey via the WeChat App.Subsequently, we obtained the responders' WeChat accounts and sent them the super link to access the multiquestionnaire survey.Before answering the survey questions, they were required to read the electronic written consent, and only those who agreed and confirmed the consent could proceed to the survey.The online survey was conducted between November 2022 and January 2023.Participation in the survey was entirely voluntary.Finally, the raw data from the survey were downloaded and analyzed.This study was approved by the Institutional Review Board of PUMCH (SK1034) and registered at Clinicaltrial.org (NCT05604716) in November 2022.

Online multiquestionnaire survey
Eight questionnaires evaluated GIF and mental stress.All questionnaires were electronically distributed using a personal online data collection tool.GIF was assessed via the Gastrointestinal Symptom Rating Scale (GSRS) and nine symptom scales of the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30 (EORTC QLQ-C30).The GSRS is a 15-item questionnaire that quantifies common GI symptoms, including abdominal pain, reflux, indigestion, diarrhea, and constipation, on a 7-point scale [16] .Higher scores indicate more severe symptoms and dysfunction.Although initially developed for patients with peptic ulcer disease and irritable bowel syndrome, it has been used in various gastrointestinal diseases and in postsurgical patients undergoing upper gastrointestinal surgery [17,18] .EORTC QLQ-C30 is a commonly used tool for assessing multiple mental stress domains, including function due to role, function due to emotional health, emotional well-being, social functioning, and general health.Each domain underwent a linear transformation to standardize the raw score on a scale of 0-100, with higher scores indicating less mental stress [19] .Social psychological functions are also critical components after surgery.We applied the depression scale of the Patient Health Questionnaire-9 (PHQ-9) and the General Anxiety Disorder 7-item Scale (GAD-7) to evaluate the degree of depression and anxiety.The Sense of Coherence scale (SOC-9) evaluated the patient's vulnerability to illness.The Post-Traumatic Embitterment Disorder self-rating scale (PTED-21) presented the intensity of reactive stimulus-bound embitterment [16,20] .Patient satisfaction with certain aspects of his or her body's appearance was scored by the Body Image Scale (BIS).The Working Ability Index (WAI) measures workers' perceptions regarding their physical, mental, and social health and ability to cope with job demands.A brief introduction and explanation of these eight scales was presented in Table 1.The super link to the online survey was https://www.wjx.cn/vm/hnHjC1U.aspx#.The whole

HIGHLIGHTS
• Solid pseudopapillary neoplasm (SPN) has extremely favorable oncological outcomes after surgery.Most patients are young or middle-aged women in their early careers.Therefore, postoperative gastrointestinal function (GIF) and mental stress of these patients are the main concerns.However, no information on GIF and the mental stress has been reported yet.We provide this information by using a multiquestionnaire survey in a 20-year cohort of SPN patients from a high-volume center.• Local resection (LR) and radical resection have been reported to have comparable oncological outcomes for SPN patients in previous studies.However, their impact on GIF and mental stress has not been reported.We now show that local resection has improved GIF and reduced mental stress compared to radical resection.• The results of this study can aid surgeons in making more precise and individualized plans for patients to meet the balance between oncological outcomes and mental stress and GIF.
questionnaire included 110 questions in total.A validated response was defined as: (1) all questions of the survey were answered; (2) the participants answered the quality-control questions correctly; (3) the time to answer the questions was more than 4 min.This work has been reported in line with the strengthening the reporting of cohort, cross-sectional, and casecontrol studies in surgery (STROCSS) criteria [21] (Supplemental Digital Content 1, http://links.lww.com/JS9/B185).

Statistical analysis
Measurement data was presented as mean SD or median with an interquartile range, and categorical data were expressed as numbers with percentages.Measurement data was compared using the independent t test or analysis of variance when the data followed a normal distribution.When the measurement data did not follow a normal distribution, the nonparametric rank-sum test was adopted.Categorical data were compared using the χ 2 test.
A multivariate linear regression model was used to explore the causal relationship and to identify independent predictors of mental stress and GIF and was described using a β coefficient and 95% CI.Correlations between mental stress and GIF were analyzed by Spearman correlation analysis.Statistical significance was set at a two-tailed P-value of <0.05.Statistical analyses were performed using the SPSS statistics software package (version 26.0; IBM).The graphs were produced using GraphPad Prism [9.5.0(525)] and SPSS.

Patient enrollment and basic information
Information on 454 SPN patients who underwent surgery from 2001 to 2021 was retrieved.A total of 352 patients responded to our telephone-call follow-up and received the online questionnaire.Eighty-five patients did not respond, and 72 patients refused to answer.We received 193 questionnaires and 183 of those could be validated (Fig. 1).The demographic and clinicalpathological characteristics of 454 cases have been reported previously [10] .The characteristics of the included 183 patients was shown in Table 2.Among them, 138 (75.4%) were female, and 65 (35.5%) were unmarried at the time of surgery.Forty-six patients (25.1%) underwent LR, while 137 (74.9%) underwent

Mental stress and GIF of the patients after surgery
The average scores for each symptom of GRSR were above the median value of the scale (3.5) in less than 20% of the patients, suggesting that most patients had satisfactory GIF (Table S1, Supplemental Digital Content 2, http://links.lww.com/JS9/B186,Fig. 2A).Sixteen patients (8.7%) showed obvious clinical symptoms of depression, while 27 patients (14.8%) showed obvious clinical symptoms of anxiety (GAD-7 and PHQ-9, ≥ 10).Nineteen (10.4%) patients complained about poor ability to work after surgery (WAI, <27.0), and 31 (16.9%) had concerns of their body image (BIS, ≥ 10).Eleven (6%) patients obtained scores that were more than half of the total score of the PTED-21 scale, while 153 (83.6%) patients scored more than half of the total score of SOC-9, indicating an excellent ability of these patients to cope with pressure (Table S2, Supplemental Digital Content 2, http://links.lww.com/JS9/B186,Fig. 2B).The results of the symptom part of EORTC QLQ-C30 showed that few patients had a score over the third quartile of all symptoms.The average insomnia score was the highest, compared to nausea and vomiting (26.6 28.3 vs. 9.9 17.5, P < 0.001) (Table S3, Supplemental Digital Content 2, http://links.lww.com/JS9/B186,Fig. 2C).Similar findings were shown in the function part of EORTC QLQ-C30.Few patients scored below the first quartile in various functions.The average value of physical function was the highest.In contrast, the average value of emotional function was the lowest (90.1 13.9 vs. 77.6 22.1, P < 0.001) (Table S4, Supplemental Digital Content 2, http://links.lww.com/JS9/B186,Fig. 2D).

Correlations between mental stress, GIF, and clinicopathological characteristics
A nonparametric sum rank test was adopted to analyze differences in the scores of mental stress and GIFs among the clinicopathological items.In general, older age patients had fewer concerns regarding their body image (P = 0.004) and worse ability to work (P = 0.035); female patients had a poorer ability to respond to stress (P = 0.021) and suffered more constipation (P = 0.015).Patients with both physical and mental work had poorer ability to work (P = 0.003) and more financial difficulty (P < 0.001).Married patients had better performance on SOC-9 (P = 0.021).The symptoms of diarrhea diminished gradually after surgery (P = 0.043).Patients with postoperative complications experienced more severe GI symptoms (P = 0.018 and P = 0.031, respectively).Unexpectedly, features of tumor invasiveness did not significantly affect mental stress and GIF.The type of surgical procedure was the dominant factor affecting mental stress and GIF.Compared to RR, LR led to reduced mental stress and substantially better GIF (Table 3 and Table 4).Regarding RR, pancreaticoduodenectomy with or without pylorus-preservation led to more severe GI symptoms, compared to distal pancreatectomy with or without splenectomy (Table S5, Supplemental Digital Content 2, http://links.lww.com/JS9/B186).
Multiple linear regression was used to analyze the correlation between mental stress, GIFs, and clinicopathological characteristics.As shown above, postoperative complications led to more severe GI symptoms.Diarrhea gradually diminished 5-years after surgery.MIS resulted in less postoperative GI dysfunction and better cognitive function (EORTC QLQ-C30, three subdomains, P < 0.05; GSRS, two subdomains and the average score, P < 0.05).LR led to less mental stress (PHQ-9, P = 0.018), better ability to work (WAI, P = 0.010), less GI dysfunction and better quality of life (EORTC QLQ-C30, nine subdomains, P < 0.05; GSRS, three subdomains and the average score, P < 0.05) (Table 5) (Table 6).

Change of marital status after surgery
A total of 65 patients were unmarried when operated upon.Up to the last follow-up, 25 (38.5%)patients were married, and 40 (61.5%) were still unmarried.Four patients got divorced after surgery.More patients under 25 years old were in the unmarried population (P < 0.001).Change in marital status was unrelated to other clinicopathological characteristics (Table S6, Supplemental Digital Content 2, http://links.lww.com/JS9/B186).

Correlations between mental stress and GIF
The correlation between each GI symptom of GSRS and mental stress was analyzed next.This analysis revealed that poorer GI symptoms had a substantial negative impact on the patients' mental stress and functions.Details were presented in Table 7.

Discussion
Increasing evidence, including our previous study, has largely alleviated concerns regarding tumor recurrence and metastasis of SPN after surgery, as LR has shown comparable oncological outcomes and morbidity rates to RR [3,4,7,22,23] .The predominant population of SPN patients comprises young and middle-aged women at a crucial period of their life and professional career.Diagnosing SPN and undergoing pancreatic surgery can potentially lead to significant mental stress and gastrointestinal dysfunction.However, no study has presented the postoperative mental stress and GIF of these patients, nor has there been any exploration of how to improve their postoperative GIF and reduce their mental stress through more precise and personalized surgical treatment strategies.In this study, we addressed these two issues by conducting a multiquestionnaire online survey in a long-term large cohort of SPN patients after surgery.
Furthermore, we innovated the assessment methods by using an online multiquestionnaire survey.We now show that 10-20% of patients had clinical symptoms of anxiety and depression after surgery, as demonstrated by poor performance on PHQ-9 and GAD-7.As indicated by WAI scores, ~10% of the patients complained of poor ability to work, and 16.9% of the patients had concerns about their body image, as indicated by BIS scores.In line, about 20% of the patients had GSRS scores above the median value.According to the EORTC QLQ-C30, the emotional function was most significantly affected compared to other functions.Mental stress and GIF correlated, indicating that more severe GI symptoms result in more mental stress.Multiple linear regression analyses showed that MIS resulted in fewer GI symptoms after surgery.It is reasonable to assume that patients with more aggressive tumors may have more severe mental stress; however, this could not be verified in the present analysis.It has to be pointed out that, since only a small proportion of patients had more aggressive tumors, this has to be interpreted carefully.Female patients complained of more severe emotional dysfunction and GI symptoms.Patients with physical work complained of more severe financial difficulties, and younger patients had more concerns regarding body image.Education level and marital status did not correlate with mental stress.
The main factor that affected mental stress and a GIF was LR.Thus, LR resulted in less mental stress and better GIF as compared to RR.The progress in pancreatic surgery and centralization in high-volume centers has made LR, including enucleation, duodenum preserving pancreatic head resection, and pancreatic segmental resection, safe and feasible [9,24,25] .During the last two decades, minimally invasive pancreatic surgery has rapidly progressed and robotic and laparoscopicassisted pancreatic procedures have become routine in highvolume centers [26][27][28] .These advanced in organ preserving and minimally invasive methods enhance the recovery of the patients and reduce operative trauma [29,30] .Multilinear regression also showed that MIS led to fewer GI symptoms after surgery.Therefore, LR and MIS can potentially improve GIF and relieve the mental stress for SPN patients undergoing surgery.

Limitations of this study
Due to the low incidence of SPN, conducting a randomized controlled study to explore the impact of LR and RR on postoperative mental stress and GIF is nearly impossible.While this study comprises a large cohort, there may be some selection biases that could potentially confound the results.In the future, external validations should be further conducted through multicenter studies in different regions and different countries to confirm the findings of this study.Additionally, being a crosssectional study, it was unable to analyze dynamic changes in mental stress and GIF over time.Therefore, a prospective study would be needed to observe the dynamic changes in mental stress and GIF of SPN patients.

Conclusion
This study presents, for the first time, data on the postoperative GIF and mental stress of SPN patients using a multiquestionnaire online survey in a large cohort from a high-volume center.The study findings demonstrate that LR and MIS significantly improve GIF and alleviate mental stress after surgery, in comparison to RR and open procedures.The results of this study could assist surgeons in devising more precise and personalized surgical plans for their patients.

Table 1
Eight questionnaires evaluating GIF and mental stress of SPN patients.
a Question No.14 of PTED-21 is not applicable to the survey.

Table 2
Clinical and surgical features of patients.

Table 3
Correlations between various subscales and clinicopathological characters.
*P-value <0.05.**P-value <0.01; b P-value was derived by the Pearson χ 2 test; the other P-values were derived by nonparametric sum rank test.

Table 4
Correlations between various subscales and surgical features.

Table 5
Multiple linear regression of subscales and clinicopathological characters.work' group as a reference to set the dummy variable.Hua et al.International Journal of Surgery (2023) a Take 'no

Table 6
Multiple linear regression of subscales and surgical features of patients.