Long-Term Follow-Up after Surgery for Chronic Pancreatitis: A Bicentric Retrospective Experience

Abstract

Introduction

The impact of surgery on nutritional status, pancreatic function, and symptoms of patients affected by chronic pancreatitis (CP) has not been unequivocally determined. This study aimed to evaluate clinical follow-up after surgery for CP in an Italian-Austrian population.

Materials and Methods

Patients operated for CP at two high-volume centers between 2000 and 2018 were analyzed. The following parameters were compared between the pre- and postoperative period: nutritional status, endocrine and exocrine pancreatic functions, and chronic pain.

Results

Overall, 186 patients underwent surgery for CP. Among these, 68 (40%) answered a specific follow-up questionnaire. The body mass index showed a significant increase between pre- and postoperative assessments (21.1 vs. 22.5 p = 0.003). Furthermore, a 60% decrease in the prevalence of chronic pain (81 vs. 21%, p < 0.001) was observed. On the contrary, both exocrine and endocrine pancreatic functions pointed toward a worsening after surgery, with consistent higher rates of patients presenting with diabetes mellitus, as well as patients requiring insulin therapy and oral intake of pancreatic enzymes. The analysis of body composition performed on 40 (24%) patients with a complete imaging pack revealed no significant change in the nutritional status after surgery.

Discussion/Conclusion

Despite the good results observed in terms of pain relief, the surgical approach led to a consistent worsening of the global pancreatic function. No significant influence of surgery on the nutritional status of patients was detected.

Introduction

Chronic pancreatitis (CP) is an enduring inflammatory disease compromising patients' quality of life, resulting in permanent structural damage of the gland [1]. The primary purpose of surgery for CP is to provide long-lasting pain relief, minimize morbidity, and maintain long-term pancreatic function [2]. However, the conventional step-up approach considers surgical intervention as the last therapeutic option both for patients with symptomatic or complicated CP undergoing conservative therapies which have been ineffective and for those with suspected pancreatic cancer [3]. Although some authors have recently promoted the early adoption of surgery − suggesting greater cost-effectiveness and efficacy as compared to endoscopy [4] − the real consequences of this approach on patients' health are still under debate [5]. Furthermore, any attempts to quantify the nutritional benefits of surgery in CP patients have not proven particularly successful [5, 6]. The lack of homogeneous data and the absence of prospective studies in the literature do not allow for an assessment of the exact prevalence of chronic malnutrition and sarcopenia − which seem to represent significantly negative prognostic factors [7] − among patients presenting with CP, as well as the repercussions of surgery on these aspects [8]. The aim of this study was to investigate the clinical effects of surgery on CP patients in a bicentric Italian-Austrian population, with a particular focus on nutritional status, pancreatic function, and chronic pain.

Materials and Methods

Study Design, Study Population, and Data Collection

This retrospective analysis included all patients undergoing surgery for CP at two university centers (Department of General Surgery of Vienna and Department of General and Pancreatic Surgery of Verona) between 2000 and 2018 (Fig. 1). The baseline Pre- and perioperative data were extracted from prospective patient registries.

Fig. 1.

Study design and inclusion and exclusion criteria.

The postoperative data were obtained using a specific questionnaire (general characteristics, pancreatic function and pain − online suppl. Fig. I; for all online suppl. material, see www.karger.com/doi/10.1159/000526263) completed between March and June 2020. Only those patients which answered the questionnaires were included in the follow-up analysis of pancreatic functions and pain, assessing and comparing the following parameters prior to and after surgery: body mass index (BMI), endocrine and exocrine pancreatic functions (oral intake of pancreatic enzymes (OPE) and prevalence of diabetes mellitus (DM) or glucose intolerance) and prevalence of chronic or recurrent pain (CRP).

The second analysis performed − namely, the imaging-based nutritional status assessment − included all patients for whom a “complete imaging pack” (a pair of pre- and postoperative radiological images of the same type; computer tomography or magnetic resonance imaging) was available. Preoperative imaging was not allowed more than 1 year prior to surgery, whereas postoperative imaging had to be performed between 1 and 4 years after surgery. Assessment was performed at the level of the third lumbar vertebra (L3) including the following parameters (online suppl. Fig. II): subcutaneous adipose tissue (SAT), total skeletal muscle mass (TSMM) (i.e., rectus abdominis, oblique and transverse abdominal muscles, paraspinal muscles, and the psoas muscle), intramuscular adipose tissue (IMAT), visceral adipose tissue (VAT), and right psoas cross-sectional area using the Slice-O-Matic® medical imaging software (Version 5.0; TomoVision, Magog, QC, Canada). All segmented images were stored anonymized and subsequently reviewed by a group of co-workers made up of two surgeons and an experienced radiologist, who checked the accuracy of the measurements and repeated the segmentation in case of ambiguous or inaccurate results.

Statistical Analysis

No sample size calculation was carried out due to the retrospective design of the study. Continuous variables are expressed as mean ± SD or as medians with interquartile range (IQR) as appropriate; categorical variables are expressed as frequencies with percentages. For categorical data, the χ² tests with Yates correction in 2-3-2 contingency tables was employed; the t Student paired test was used to compare mean values, whereas the Wilcoxon test was used to compare medians. To allow for different time points in postoperative imaging, a linear regression model was used to analyze the effect of time on outcome measures. In general, a two-sided p value <0.05 was considered to be statistically significant. The statistical analysis was conducted by using SPSS Statistics software version 26.0 (IBM Corporation, Armonk, NY, USA) and Stata 14.0 (Stata Corp, College Station, TX, USA).

Results

Between 2000 and 2018, a total of 186 patients (77 from Vienna, 91 from Verona) received surgery for CP at the two centers. Following the exclusion of 18 patients, 168 fulfilled the baseline eligibility criteria of the study. Nevertheless, only 68 patients returned the mailed questionnaires and could be included in the analysis of pancreatic functions and pain, while the body composition assessment was performed on the 40 patients with a complete imaging pack (Fig. 1).

The Italian and Austrian patients displayed strong similarities in terms of general baseline characteristics and pathological features at the time of diagnosis (online suppl. Table I). The patient's baseline characteristics and the data relating to the perioperative course are shown in Table 1. Most of the included patients were men (80%) aged around 50 years with a median Charlson age-adjusted comorbidity index of 1.5 (IQR 2). The most relevant preoperative risk factor was cigarette smoking (69%), and the most frequently described symptom was CRP (78%). Almost 50% of patients had undergone one or more (median 2, IQR 2.25) preoperative attempts at interventional endoscopy. The median time-lapse between diagnosis and surgery was three (IQR 5) years, and most surgical operations − resective (80%) or decompressive (20%) − were performed by using an open technique (99%). The overall rates of postoperative morbidity and mortality were 33% and 1%, respectively.

Table 1.

General features (n = 168)

Sex, n (%)
M	135 (80)
F	33 (20)
Age at diagnosis, mean (±SD), years	49.7 (±11.48)
ACCI, median (IQR)	1.5 (2)
BMI, median (IQR)	21.5 (4.4)
Smoking habit, n (%)	116 (69)
Habitual alcohol consumption, n (%)	78 (46)
Familiarity for CP, n (%)	9 (5)
Main clinical symptoms, n (%)
Recurrent or chronic pain	131 (78)
Acute on CP	68 (40)
Weight loss	57 (34)
Loss of appetite	27 (16)
Dyspepsia	24 (14)
Preoperative endoscopy, n (%)	81 (48)
Preoperative endoscopies, median (IQR), n	2 (2.25)
Surgical indication/clinical suspect, n (%)
CP	147 (88)
Groove pancreatitis	18 (11)
PDAC	2 (1)
Pancreatic cyst(s)	1 (1)
Time lapse (first diagnosis to surgery), median
(IQR), years	3 (5)
Type of surgical intervention, n (%)
Distal pancreatectomy	25 (15)
PPPD	87 (52)
Whipple PD	21 (13)
Partington-Rochelle	17 (10)
Frey	8 (5)
Total pancreatectomy	3 (2)
Other	5 (3)
Open versus VLS, n (%)
Open	167 (99)
VLS	1 (1)
Postoperative complications (overall), n (%)	56 (33)
POPF, n (%)
BL	10 (6)
B	4 (2)
C	8 (5)
DGE, n (%)	3 (2)
PPH, n (%)
A	10 (6)
B	3 (2)
C	0
Biliary fistula, n (%)	6 (4)
Other complications, n (%)	49 (29)
Reintervention, n (%)	10 (6)
30-day mortality, n (%)	1 (1)
Hospital stay, median (IQR), days	9 (7)

SD, standard deviation; IQR, interquartile range; ACCI, age-adjusted Charlson comorbidity index; BMI, body mass index; PDAC, pancreatic ductal adenocarcinoma; PPPD, pylorus-preserving pancreatoduodenectomy; VLS, videolaparoscopy; POPF, post-operative pancreatic fistula; BL, biochemical leak; DGE, delayed gastric emptying; PPH, post-pancreatectomy hemorrhage.

Pancreatic Function and Pain

Overall, 68 (40%) patients answered the questionnaire after a median timeframe of 89 months (IQR 72). Among the responders, 15% (10 of 68) were still smokers at the timepoint of follow-up, while 4% (three of 68) stated active and regular alcohol consumption. After surgery (Table 2), patients showed a significant increase in BMI (21.1 vs. 22.5, p = 0.003), as well as a reduction in CRP (81 vs. 21%, p < 0.001). Consequently, most did not require long-term analgesics at the time of follow-up. Conversely, surgery led to a worsening of pancreatic endocrine and exocrine functions (Table 2) in terms of a higher prevalence of DM (50 vs. 13%, p < 0.001) and a more frequent need for chronic antidiabetic therapy (AT) (50 vs. 10%, p < 0.001) and OPE (81 vs. 34%, p < 0.001).

Table 2.

Impact of surgery on BMI, pancreatic functions, and pain (n = 68)

	Pre-op	Post-op	p value
BMI, median (IQR)	21.1 (4.4)	22.5 (5)	0.003
DM, n (%)	9 (13)	34 (50)	<0.001
DM therapy, n (%)
No therapy	61 (90)	34 (50)	<0.001
Oral antidiabetics	2 (3)	11 (16)
Insulin	5 (7)	21 (31)
Both	0	2 (3)
Enzymes (oral intake), n (%)	23 (34)	55 (81)	<0.001
Recurrent or chronic pain, n (%)	55 (81)	14 (21)	<0.001
Chronic intake of analgesic therapy, n (%)
No	33 (49)	65 (96)	<0.001
NSAIDs	11 (16)	2 (3)
Opioids	15 (22)	1 (1)
Drugs combination	9 (13)	0

BMI, body mass index; DM, diabetes mellitus; IQR, interquartile range; NSAIDs, nonsteroidal anti-inflammatory drugs.

Imaging Analysis

A total of 40 patients (24%) with a “complete imaging pack” showed no change in body composition parameters measured before and after surgery (Table 3). The linear regression analysis revealed no effect of time (median timespan between images 24 months, IQR 30) on outcome variables (SAT, TSMM, IMAT, VAT, and right psoas cross-sectional area) (online suppl. Table II).

Table 3.

Impact of surgery on nutritional status (n = 40)

	Pre-op	Post-op	Δ(postop − preop)	p value
TSMM, mean ± SD, cm2	137.2±33.4	134.8±30.9	−2.42±16.43	0.931
Psoas area, mean ± SD, cm2	8.1±2.7	8.0±2.6	−0.85±1.85	0.290
IMAT, mean ± SD, cm2	9.2±4.8	10.5±5.8	+ 1.21±4.39	0.754
VAT, mean ± SD, cm2	77.7±50.2	77.8±62.6	+ 1.72±62.03	0.170
SAT, mean ± SD, cm2	99.5±63.4	99.6±60.7	+0.65±48.92	0.080

Δ^{(postop − preop)} = difference between post- and preoperative measured values; TSMM, total skeletal muscle mass; IMAT, intramuscular adipose tissue; VAT, visceral adipose tissue; SAT, subcutaneous adipose tissue.

Discussion

Here, we present the follow-up data from 2 decades of bicentric experience in surgery for CP. According to our findings, surgery appears to offer no substantial benefit regarding the nutritional status and body composition. In addition, surgery seems to a negative effect on both endocrine and exocrine pancreatic functions, leading to a postoperative increase in the rate of DM, as well as the need for AT and OPE. By contrast, a considerable decrease in the prevalence of CRP was observed between the pre- and postoperative period.

CP has potentially detrimental effect on the nutritional status of patients. It has been estimated that between 20% and 70% of patients affected by CP are underweight or malnourished [8]. This impairment of the nutritional status has been proven to be associated with increased morbidity and mortality [2]. Sarcopenia, in particular, has been recognized as being associated with a significant reduction in QoL, increased hospitalization, and reduced survival [7]. Interestingly, our patients presented with a median preoperative BMI in the normal range, according to the World Health Organization (WHO) referral values [9] (Table 1). This finding supports our observations in clinical practice, namely, that advanced stages of physical consumption in CP are becoming increasingly rare. This phenomenon could not only be explained by a lower frequency of exclusively alcohol-based feeding regimes among patients suffering from CP [10] but also by the radical disruption in dietary habits occurring in recent decades in favor of fatty, high-calorie, and nutrient-rich foods [11]. Our analysis showed an increase in BMI between the pre- and postoperative assessments (from 21.1 to 22.5, p = 0.003), which, at first glance, might be viewed as a positive consequence of surgery. Nevertheless, preoperative BMI has been shown to be a limited marker of malnutrition and does not accurately predict the development of sarcopenia [12]. This is why many authors have endeavored to find simple, reliable, and feasible imaging-based methods to define body composition in individuals affected by CP [13]. Specifically, Bieliuniene et al. have demonstrated a high prevalence of sarcopenia among CP and pancreatic ductal adenocarcinoma patients by using a computer tomography- and magnetic resonance imaging-based assessment of body composition [14]. Based on the aforementioned experiences and on the imaging methods described in the literature [15], we performed an assessment of the nutritional status of patients in our cohort before and after CP surgery. As a result, no relevant variation in any parameter measured (TSMM, IMAT, VAT, SAT, psoas area) was observed: surgery did not seem to have a consistent impact on the body composition when based on imaging comparison (Table 3). Incidentally, only those measures representing the body fat distribution (VAT, SAT, IMAT) showed a positive (although not significant) trend in the inter-observational period. Despite the paucity of data and the unique nature of such an analysis, this could be interpreted as evidence of the complexity of the metabolic changes occurring in CP: in particular, the muscle mass loss and sarcopenia seem to be chronic processes that require longer time to onset but, at the same time, prove to be more refractory to therapeutical interventions and tend to become irreversible over time.

The real effect of CP surgery on exocrine and endocrine pancreatic functions is still largely unknown. Although studies examining the impact of surgical drainage or resection on endocrine insufficiency yielded little or no results [16, 17], many authors have suggested that the early adoption of surgery could prevent the development of exocrine insufficiency [18, 19]. Our findings showed a global worsening of both pancreatic exocrine and endocrine functions after surgery. On the other hand, the median timeframe between onset of symptoms and surgery in our series was 3 (IQR 5) years. Therefore, it is not possible to determine whether the results obtained represent the direct consequence of surgical intervention or if they reflect the non-adoption of an early surgery policy. However, the most widely acknowledged benefit to be derived from CP surgery is chronic pain relief [19], as substantiated by our findings. Indeed, Table 2 indicates a before-after surgery reduction (approx. 60%) in the number of patients suffering from CRP, as well as a decreased intake of analgesics. Current evidence points toward the beneficial role of surgery in patients with painful CP when performed within 2–3 years from the initial diagnosis and after a maximum of 5 attempts at endotherapy [5]. Nevertheless, this positive effect relating to pain relief should be interpreted with some caution since most studies are based on retrospective uncontrolled surgical series and may therefore overestimate the actual impact of surgery. Many experiences in the literature [20, 21] have suggested that surgery as a first approach in patients with symptomatic obstructive CP has proved to be more efficient and effective when compared to the traditional step-up policy in relieving obstruction, achieving long-term pain relief and improving patients' QoL [21, 22]. However, it is worth pointing out that surgery is burdened with a non-negligible rate of postoperative morbidity and mortality (in our series, 33% and 1%, respectively). Moreover, endoscopy has proved highly successful in patients with uncomplicated, short-term CP and occasional acute pain [23]. Hence, great caution is needed when considering the concept of “early surgery.” In the meantime, selection of the best treatment should always be guided by a careful risk-benefit assessment and by a multidisciplinary evaluation [24].

By design, the retrospective nature of the study leads to a limited interpretation of our findings. Despite the efforts made to minimize potential confounders affecting the reliability of our analysis (e.g., by excluding pancreatic ductal adenocarcinoma, autoimmune pancreatitis, and systemic inflammatory diseases), many limiting factors should be mentioned. The assessment of postoperative BMI, exocrine and endocrine pancreatic function, pain, and analgesic therapy was based on a questionnaire and therefore dependent on truthful and medically unsupervised answers. Moreover, a direct correlation between clinical and radiological parameter could not be performed since only few patients with a complete imaging pack answered the abovementioned questionnaire (Fig. 1); for this reason, the analyses were substantially conducted on two different sub-cohorts. Another relevant limitation was the incomplete clinical information, including the unavailability of many specific laboratory parameters (e.g., fecal fat globulin, glycated hemoglobin, insulin, and C-peptide levels): due to the lack of data, it was not possible to precisely quantify the degree of exocrine and endocrine pancreatic damage. Furthermore, the long and varying timespan between surgery und follow-up, together with a limited response rate, may have influenced our results. Lastly, the reliability of the imaging analysis was, to a certain degree, limited by the small number of patients (n = 40, 24%) with a complete imaging pack.

Conclusion

Surgery for CP did not show any effect on the variations in body composition and nutritional status of patients in our cohort, suggesting that these aspects are part of a complex and multifactorial process that deserves further in-depth inquiry. In spite of its high effectiveness in terms of long-lasting pain relief, the surgical approach resulted in a consistent worsening of endocrine and exocrine pancreatic functions.

Statement of Ethics

This study was approved by the Ethics Committee of the University of Vienna (protocol No. 1153/2020) and the Ethics Committee for Clinical Research of the provinces of Verona and Rovigo (protocol No. 1101CESC) and was performed in compliance with the Good Clinical Practice standards and the principles of the Declaration of Helsinki. A written informed consent was obtained from all individual participants included in the study.

Conflict of Interest Statement

The authors have no conflict of interest to declare.

Funding Sources

The authors received no financial support for the research, authorship, and/or publication of this article.

Author Contributions

Study conception and design: Niccolò Surci, Jakob Mühlbacher, and Klaus Sahora. Data acquisition: Niccolò Surci, Giacomo Deiro, Christina Bergmann, and Klaus Walenta. Data analysis and interpretation: Niccolò Surci, Jakob Mühlbacher, Giovanni Marchegiani, and Klaus Sahora. Drafting of manuscript: Niccolò Surci, Jakob Mühlbacher, and Giovanni Marchegiani. Critical revision of the manuscript: Claudio Bassi, Roberto Salvia, Klaus Sahora, Luca Casetti, Dietmar Tamandl, and Martin Schindl.

Data Availability Statement

The data that support the findings of this study are not publicly available because they contain information that could compromise the privacy of research participants but are available upon request from Jakob Mühlbacher.

Supplementary Material

Supplementary data

Funding Statement

The authors received no financial support for the research, authorship, and/or publication of this article.