You are here

  • Home
  • Online First
  • Early postoperative pain and 30-day complications following major abdominal surgery: a retrospective cohort study

Article Text

Article menu
Regional anesthesia and acute pain
Early postoperative pain and 30-day complications following major abdominal surgery: a retrospective cohort study
  1. Esmee van Helden1,
  2. Josephine Kranendonk1,
  3. Ad Vermulst2,
  4. Arjen de Boer1,
  5. Philip de Reuver1,
  6. Camiel Rosman1,
  7. Johannes de Wilt1,
  8. Kees van Laarhoven1,
  9. Gert Jan Scheffer1,
  10. Christiaan Keijzer1 and
  11. Michiel Warlé1
  1. 1Radboudumc, Nijmegen, Gelderland, The Netherlands
  2. 2Mental Healthcare East-Brabant Region Helmond—Peelland, Boekel, Oost-Brabant, The Netherlands
  1. Correspondence to Dr Esmee van Helden, Radboudumc, Nijmegen 6500 HB, The Netherlands; esmee.vanhelden{at}radboudumc.nl

Abstract

Background Increasing evidence supports a positive relationship between the intensity of early postoperative pain, and the risk of 30-day postoperative complications. Higher pain levels may hamper recovery and contribute to immunosuppression after surgery. This leaves patients at risk of postoperative complications.

Methods One thousand patients who underwent major abdominal surgery (cytoreductive surgery and hyperthermic intraperitoneal chemotherapy, esophageal, liver, or pancreas surgery) at the Radboud university medical center were randomly selected from eligible patients between 2014 and 2020. Pain scores on day 1, the independent variable of interest, were extracted from the electronic patient files. Outcome measures were 30-day postoperative complications (infectious, non-infectious, total complications and classification according to Clavien-Dindo).

Results Seven hundred ninety complications occurred in 572 patients within 30 days after surgery, of which 289 (36.7%) were of infectious origin, and 501 (63.4%) complications were non-infectious. The mean duration from the end of surgery to the occurrence of infectious complications was 6.5 days (SD 5.6) and 4.1 days (SD 4.7) for non-infectious complications (p<0.001). Logistic regression analysis revealed that pain scores on postoperative day 1 (POD1) were significantly positively associated with 30-day total complications after surgery (OR=1.132, 95% CI (1.076 to 1.190)), Clavien-Dindo classification (OR=1.131, 95% CI (1.071 to 1.193)), infectious complications (OR=1.126, 95% CI (1.059 to 1.196)), and non-infectious complications (OR=1.079, 95% CI (1.022 to 1.140)).

Conclusions After major abdominal surgery, higher postoperative pain scores on day 1 are associated with an increased risk of 30-day postoperative complications. Further studies should pursue whether optimization of perioperative analgesia can improve immune homeostasis, reduce complications after surgery and enhance postoperative recovery.

  • COMPLICATIONS
  • Pain, Postoperative
  • Postoperative Complications
  • Pain Measurement

Data availability statement

Data are available upon reasonable request.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, an indication of whether changes were made, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

https://doi.org/10.1136/rapm-2024-105277

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    WHAT IS ALREADY KNOWN ON THIS TOPIC

    • It is well documented that early postoperative pain can lead to less-favorable patient outcomes, including decreased satisfaction, increased incidence of chronic pain, and extended hospital stays, all of which contribute to a greater likelihood of readmission.

    WHAT THIS STUDY ADDS

    • This study sheds light on the significant correlation between high pain scores on the first day after surgery and the incidence of both infectious and non-infectious complications within the first 30 days post surgery. This emphasizes the importance of managing early postoperative pain effectively to potentially mitigate complications after major abdominal surgery.

    HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

    • The study underscores the critical role of enhanced analgesic strategies in elective major abdominal surgeries, aiming to improve patient comfort and expedite recovery. This could potentially translate into a decrease in postoperative complications.

    Introduction

    Early postoperative pain is associated with lower patient satisfaction,1 delayed postoperative ambulation, the development of chronic pain,2 prolonged length of hospital stay, and a higher readmission rate.1 Moreover, experiencing postoperative pain is the most common patient concern.3 There is an increasing body of evidence that supports a positive relationship between the intensity of early postoperative pain and postoperative complications, especially infectious complications.4–7 More complex and extensive surgical procedures are associated with higher levels of postoperative pain and a higher rate of postoperative complications.4 7–10 The central role of acute pain in the origin of postoperative complications has been broadly accepted and forms the main rationale for comprehensive pain management in the early phase after surgery. Postoperative 30-day complications, including moderate Clavien-Dindo (CD) grade 2 complications, have a significant impact not only on healthcare costs11 but also on long-term survival outcomes after curative oncological surgery.12 13

    Pain control has become an integral part of the enhanced recovery after surgery protocols.5 Despite a broad implementation of pain management protocols for many care paths, and a strong involvement of the acute pain service in our hospital, we previously reported a relatively high incidence of unacceptable postoperative pain in a broad surgical population.5 Early detection of unacceptable postoperative pain is imperative for adequate pain management after surgery. Especially in major abdominal surgery, the importance of adequate recognition and treatment of severe and/or unacceptable pain after surgery is widely acknowledged by clinicians. Considering the extensiveness of these surgeries and relatively high incidence of postoperative complications14–16 compared with other abdominal procedures, assessment of the relationship between early postoperative pain and postoperative complications in this population can prove valuable to further identify patients where individualized and comprehensive pain management are extra important to enhance patient outcomes. In this retrospective cohort study, we assess the relationship between the intensity of early postoperative pain and the risk of infectious and non-infectious complications within 30 days after major abdominal surgery.

    Methods

    Patients

    The medical database of the Radboud university medical center was used to extract data for our study. All adult patients who underwent elective major abdominal surgery between January 2014 and December 2020 were considered eligible for participation in our study. There was no preselection of patients to be included. Major abdominal procedures included were cytoreductive surgery (CRS) and hyperthermic intraperitoneal chemotherapy (HIPEC), esophageal, liver and pancreatic surgery. From a total of 1659 eligible patients, a subgroup of 1000 patients was randomly selected by electronic Simple Random Sampling to minimize bias and allow generalization of findings.

    Independent variable: postoperative pain scores on day 1

    Postoperative pain scores were extracted from the electronic patient files on postoperative day (POD) 1 on a Numeric Rating Scale (NRS) from 0 (no pain) to 10 (worst possible pain).

    Covariates: baselineoperative, intraoperative and postoperative variables

    Collected baseline, intraoperative and postoperative variables were gender, age, body mass index (BMI), American Society of Anesthesiologists (ASA) score, perioperative epidural or paravertebral catheters, type of surgery, surgical technique (open or laparoscopic) and duration of surgery. Type of surgery (CRS and HIPEC, esophageal, liver or pancreas surgery) is a multinomial variable; the number of patients in each category will be different. Therefore, weighted effect coding will be applied with CRS and HIPEC as reference category.

    Outcome variables: 30-day postoperative complications

    Primary outcome variables were 30-day postoperative infectious and non-infectious complications (CD grade ≥2), scored as binary (0/1) variables. Infectious complications were scored according to the Centers for Disease Control definitions of healthcare-associated infections.

    Secondary outcome variables were the 30-day postoperative total complication score and postoperative complications according to the validated CD classification. The total complications score is the sum of the two primary outcome variables resulting in three possible scores: 0 (no complications), 1 (infectious or non-infectious complication) and 2 (infectious and non-infectious complication). The validated CD classification is widely used to assess and report postoperative complications in general surgery.17–20 The classification consists of five grades depending on the need for treatment and degree of threat to life varying from grade 1 complications (any deviation from the normal postoperative course without the need for intervention) to grade 5 (death of a patient).18 A researcher not involved in the patient’s care extracted complications from the electronic patient file and classified them according to the CD classification. Complications occurring in the operating room and complications directly related to anesthesia were not included. Only CD grade 2 or higher complications were taken into account to avoid bias related to under-reporting of mild CD grade 1 complications. In patients with more than one grade 2 or higher complication, only the most severe infectious and/or non-infectious complication was recorded. Therefore, a maximum of two complications per patient was registered.

    Statistical analyses

    Logistic regression analysis was used to test our hypotheses with regard to postoperative infectious and non-infectious complications (binary variables), total complications and CD classification (ordinal variables) with the statistical program Mplus V.7.2.21 Effect sizes or reported explained variances of logit regression models of the dependent variable are explained variances of the logit of the dependent variable. Values of 0.01 are small, 0.09 medium and 0.25 large.22 For the ordinal dependent variables, it is assumed that the regression lines are parallel for each of the values of the ordinal variables. We tested this proportional odds assumption with the Wald χ2 test. To assess each covariate’s unique impact on the outcomes, we initially considered them separately. Then, to control for confounding and clarify our focal variables' effects, we incorporated all covariates into a simultaneous analysis.

    In our analysis, we identified six patterns of missing data and addressed them using the full information maximum likelihood (FIML) estimator, under the assumption that data are missing at random.23 This method is standard in Mplus, yielding results comparable to multiple imputation.24 25

    The pain score from POD1, though ordinal, was used as a continuous predictor due to its nearly normal distribution and a linear relationship with the logits of our dependent variables, confirmed by the Box-Tidwell test26 27 and visual checks of scatter diagrams.

    Results

    Patient characteristics

    Patients had a mean age of 63±11 years; 58% were men. Table 1 shows the characteristics of all patients, and categorized per type of surgery. The mean duration of surgery was 5 hours and 29 min. The majority of surgeries (72%) were performed by laparotomy. An epidural catheter was placed in 89% of patients.

    View this table:
    Table 1

    Baseline characteristics

    From the random sample of 1000 patients, we found incomplete records for BMI in 141 cases, ASA classification in 23 cases and pain scores on POD1 in 133 cases. The approach and data regarding these missing values are described extensively in the online supplemental materials.

    Supplemental material

    Pain intensity

    Table 2 shows pain scores on POD1 on the NRS score from 0 to 10 per type of surgery. Pain scores on POD1 were recorded in 867 patients (87%). The mean NRS on POD1 was 5.3 for patients who underwent CRS and HIPEC, 5.0 for esophageal surgery, 4.1 for liver surgery 4.1 and 5.1 for pancreatic surgery.

    View this table:
    Table 2

    Pain intensity (NRS) at postoperative day 1

    30-day postoperative complications

    A total of 790 postoperative complications CD grade 2 or higher occurred within 30 days after surgery in 572 patients, of which 289 were infectious and 501 were non-infectious (table 3). The highest incidence of infectious complications was found after esophageal surgery in 38.5% of patients. The highest incidence of non-infectious complications occurred after pancreatic surgery in 62.9% of patients. In patients with multiple CD grade ≥2 complications, only the most severe infectious and/or non-infectious complication was scored. Thirty-five per cent of all patients had one 30-day CD grade ≥2 complication, and 22% had two CD grade ≥2 complications. A CD grade 3 complication was observed in 11% of patients, whereas a CD grade 4 or 5 complication occurred in 7.9% of patients.

    View this table:
    Table 3

    Total complications 30 days postoperative Clavien-Dindo (CD) grade 2 or higher

    Figure 1 shows the time between surgery until occurrence of infectious and non-infectious complications. The mean duration from the end of surgery to occurrence of infectious complications was 6.5±5.6 days; non-infectious complications were diagnosed after a mean of 4.1±4.7 days, a significant difference (Wald’s test, z=3.62, p<0.001).

    Figure 1
    Figure 1

    Time in days from surgery to occurrence of infectious and non-infectious complications.

    Predictors of the occurrence and severity of postoperative complications

    First, pain POD 1 and each of the covariates were used as an individual estimator of the dependent variable in a logistic regression model (online supplemental table 1). This analysis indicated that pain intensity at POD1 was significantly associated with postoperative infectious, non-infectious and total complications, and with complication severity as reflected by the CD grade. The explained variances of the logits of these four dependent variables by only pain POD1 were 0.039, 0.043, 0.039 and 0.020, to be interpreted as small effect sizes. Next, Pain POD 1 and all covariates were combined in logistic regression models as predictor of the dependent variables (table 4).

    Supplemental material

    View this table:
    Table 4

    Adjusted association between postoperative pain and complications

    For postoperative total complications, the following covariates were significant: age (OR=1.019, 95% CI (1.008 to 1.031)), ASA class (OR=1.723, 95% CI (1.401 to 2.121)), esophageal surgery (OR=1.626, 95% CI (1.245 to 2.125)), liver surgery OR=0.725, 95% CI (0.561 to 0.938)), open procedure (OR=2.174, 95% CI (1.505 to 3.139)), duration of surgery (OR=1.230, 95% CI (1.139 to 1.330)), and pain score on POD1 (OR=1.132, 95% CI (1.076 to 1.190)). All predictors together explained 21.4% of the variance of the logit of postoperative complications, a medium-to-large effect size. In addition, the pain score on POD1 was a significant predictor of total complications arising after POD1 (OR=1.116, 95% CI (1.053 to 1.183)).

    For CD classification, significant predictors were age (OR=1.018, 95% CI (1.006 to 1.031)), ASA classification (OR=1.563, 95% CI (1.266 to 1.930)), esophageal surgery (OR=2.261, 95% CI (1.650 to 3.099)), liver surgery (OR=0.732, 95% CI (0.561 to 0.956)), open procedure (OR=2.673, 95% CI (1.744 to 4.097)), duration of surgery (OR=1.280, 95% CI (1.166 to 1.405)), and the mean pain score on POD1 (OR=1.131, 95% CI (1.071 to 1.193)). All predictors together explained 21.6% of the variance of the logit of CD classification, a medium-to-large effect size. The pain score on POD1 remains a significant predictor of CD classification after POD1 (OR=1.116, 95% CI (1.052 to 1.184)).

    Postoperative infectious complications were significantly associated with age (OR=1.016, 95% CI (1.002 to 1.031)), BMI (OR=1.042, 95% CI (1.007 to 1.078)), ASA class (OR=1.674, 95% CI (1.303 to 2.151)), esophageal surgery (OR=1.856, 95% CI (1.361 to 2.531)), liver surgery (OR=0.614, 95% CI (0.455 to 0.829)), open procedure (OR=1.632, 95% CI (1.051 to 2.534)), duration of surgery (OR=1.099, 95% CI (1.015 to 1.191)), and the pain score on POD1 (OR=1.126, 95% CI (1.059 to 1.196)). All predictors together explained 17.7% of the variance of the logit of postoperative infectious complications. Pain on POD1 was also associated with complications arising after POD1 (OR=1.108, 95% CI (1.034 to 1.187)).

    Non-infectious complications were significantly associated with age (OR=1.018, 95% CI (1.005 to 1.032)), ASA (OR=1.335, 95% CI (1.060 to 1.680)), esophageal surgery (OR=1.593, 95% CI (1.174 to 2.161)), open surgery (OR=2.317, 95% CI (1.507 to 3.562)), duration of surgery (OR=1.215, 95% CI (1.117 to 1.321)) and the pain score on POD1 (OR=1.079, 95% CI (1.022 to 1.140)). All predictors together explained 15.9% of the variance of the logit of postoperative other complications. The pain score on POD1 was also associated with complications arising after POD1 (OR=1.073, 95% CI (1.007 to 1.144)).

    Discussion

    This study revealed a positive significant association between the pain scores on POD1 and CD grade 2 or higher overall, infectious, and non-infectious complications after major abdominal surgery. This association remained significant for complications arising after POD1.

    These results are in accordance with our previous findings regarding complications after laparoscopic donor nephrectomy and in a broad surgical population.4 5 Intriguingly, the results in this population further illustrate the two previously proposed underlying pathways. Complications might be mediators in causal pathways linking baseline preoperative factors (eg, type of surgery), perioperative factors (eg, duration of surgery) to postsurgical pain, and could initiate a direct causal impact on postsurgical pain.28 To address this type of bias, a logit regression analysis was performed in which pain on POD1 was related only to complications occurring after POD1. This relationship was significant for all complications. While of course postoperative complications often contribute to higher levels of postoperative pain,29 the other way around also holds true. Logically, extensive early postoperative pain directly hampers recovery as it impairs mobilization, delays removal of intravenous lines and urine catheters and leaves patients prone to catheter-related infections and respiratory infections due to superficial respiration.4 5 30 Oral intake and bowel function after surgery relate to both postoperative pain and the side effects of analgesia, mainly opioids. The mean duration from end of surgery to the diagnosis of infectious complications was 6.5 days on average and 4.1 days for non-infectious complications. Separate analysis of only the complications arising after POD1 further stress the significance of early pain intensity in development of the infectious and non-infectious complications.

    Postoperative pain contributes to immune suppression, and thereby leaves patients more susceptible to postoperative infections.30 It is well known that pain causes a physical stress reaction, reflected by activation of the hypothalamic–pituitary–adrenal axis, resulting in the release of hormones such as cortisol31 that directly affect the immune response.32 Moreover, tissue injury caused by surgery stimulates afferent neurons (AE). Distinct from the parasympathetic nervous system, the sympathetic nervous system (SNS) is crucial in the regulation of immune responses.33 Both central immune organs, which include the bone marrow and thymus, and peripheral immune organs like the lymph nodes, spleen and mucosal-associated lymphoid tissue are innervated by the SNS, and the immune cells with adrenoceptors may be regulated by norepinephrine (NE) released from the sympathetic postganglionic neurons, causing adrenergic stress-induced immunosuppressive effects. The locus coeruleus–noradrenergic system contains NE-synthesizing neurons that send diffuse projections throughout the central nervous system. The LC-NE system has a major role in arousal, attention and stress responses.34 35

    The reported prevalence of early moderate-to-severe postoperative pain (defined as NRS >4) in the international PAIN OUT, the first international acute pain registry of acute postoperative pain is 70%.36 In 2019, we reported a prevalence of 55% for moderate to severe pain (NRS >4) on POD1 in a broad surgical population.5 For all four major abdominal surgical procedures investigated in this study (CRS and HIPEC, esophageal surgery, liver surgery and pancreatic surgery), mean pain scores on POD1 were above NRS 4. While differences in definition of pain intensity, comorbidities, classification systems and methodology used may impede comparison between studies,5 37 it is well established that effective postoperative analgesia and adequate pain control for these major abdominal procedures remain a challenge for both surgeons and anesthesiologists.38 39 A considerable number of patients (79%) experienced a CD grade 2 or higher complication within 30 days after surgery, of which 29% were infectious and 50% were non-infectious complications. These numbers correspond with reported incidences of 30-day postoperative complications after these major abdominal procedures in other trials.14–16

    A strength of this study is the relatively large random sample of a cohort of major abdominal surgeries within a large academic teaching hospital. Pain scores were recorded prospectively and complications were retrieved from the electronic patient files in a retrospective manner by researchers blinded to the pain scores to minimize the risk of selection and observer bias. Moreover, clear definitions were used to assess infectious complications and complication severity (CD).11–13 17

    Our study has some limitations. One limitation is that the study population is heterogeneous with four different types of surgeries and different surgical approaches. After major complex surgery, both complication rates and pain levels are expected to be higher40 and therefore a certain degree of confounding by case-complexity cannot be ruled out. In the logit regression analyses, all available covariates related to the complexity of surgery were included. Despite this, the absence of detailed variables regarding patients’ pre-existing pain and comprehensive comorbidity profiles means that some level of residual confounding is an inevitable aspect of our analysis. We determined the variance explained by our model for the logit of postoperative complications to be 21.4%, which we interpret as a medium-to-large effect size based on Cohen’s framework. However, it’s important to note that this still leaves a significant portion of variance unaccounted for. This unexplained variance highlights areas for future research.

    Another limitation of this study is under-reporting of minor grade I complications. These minor adverse events are highly prone for under-reporting as healthcare professionals may not report them because of their small therapeutic consequences. In general, minor grade 1, complications are considered less relevant regarding healthcare costs and patients satisfaction when compared with moderate/severe grade complications.11 12 41

    A final limitation concerns the possible effect of missing pain scores on POD1. In the random sample of 1000 patients, we encountered missing data for BMI (n=141), ASA (n=23) and pain scores on POD1 (n=133). Missing data were treated with the FIML estimator.24 25 Despite the rigorous application of FIML, we acknowledge the possibility that missing data could have some impact on our findings.

    Our findings confirm the association between higher postoperative pain levels and the risk of infectious and non-infectious complications after major abdominal surgery. This implicates the need for further optimization of analgesia during and after surgery to maintain immune homeostasis and enhance postoperative recovery.

    Data availability statement

    Data are available upon reasonable request.

    Ethics statements

    Patient consent for publication

    Ethics approval

    The Institutional Review Board of Radboud University Medical Center (Nijmegen, The Netherlands) approved the study and waived the requirement for additional written informed consent. Radboudumc is a university hospital; all patient data are collected in the electronic medical record system where patients can register consent for the use of medical information for trial purposes. This retrospective cohort study was registered at clinicaltrials.gov under trial number NCT05244655.

    References

    1. ↵
      1. Regenbogen SE,
      2. Mullard AJ,
      3. Peters N, et al
      . Hospital analgesia practices and patient-reported pain after colorectal resection. Ann Surg 2016;264:1044–50. doi:10.1097/SLA.0000000000001541
      OpenUrlPubMed
    2. ↵
      1. Kehlet H,
      2. Jensen TS,
      3. Woolf CJ
      . Persistent postsurgical pain: risk factors and prevention. Lancet 2006;367:1618–25. doi:10.1016/S0140-6736(06)68700-X
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Apfelbaum JL,
      2. Chen C,
      3. Mehta SS, et al
      . Postoperative pain experience: results from a national survey suggest postoperative pain continues to be undermanaged. Anesth Analg 2003;97:534–40. doi:10.1213/01.ANE.0000068822.10113.9E
      OpenUrlCrossRefPubMedWeb of Science
    4. ↵
      1. Albers KI,
      2. van Helden EV,
      3. Dahan A, et al
      . Early postoperative pain after laparoscopic donor nephrectomy predicts 30-day postoperative infectious complications: a pooled analysis of randomized controlled trials. Pain 2020;161:1565–70. doi:10.1097/j.pain.0000000000001842
      OpenUrlCrossRef
    5. ↵
      1. van Boekel RLM,
      2. Warlé MC,
      3. Nielen RGC, et al
      . Relationship between postoperative pain and overall 30-day complications in a broad surgical population: an observational study. Ann Surg 2019;269:856–65. doi:10.1097/SLA.0000000000002583
      OpenUrlPubMed
    6. ↵
      1. Beattie WS,
      2. Badner NH,
      3. Choi PT-L
      . Meta-analysis demonstrates statistically significant reduction in postoperative myocardial infarction with the use of thoracic epidural analgesia. Anesth Analg 2003;97:919–20. doi:10.1213/01.ANE.0000074660.05733.10
      OpenUrlPubMedWeb of Science
    7. ↵
      1. Rodgers A,
      2. Walker N,
      3. Schug S, et al
      . Reduction of postoperative mortality and morbidity with epidural or spinal anaesthesia: results from overview of randomised trials. BMJ 2000;321:1493. doi:10.1136/bmj.321.7275.1493
      OpenUrlAbstract/FREE Full Text
    8. ↵
      1. Bruintjes MHD,
      2. van Helden EV,
      3. de Vries M, et al
      . Chronic pain following laparoscopic living-donor nephrectomy: prevalence and impact on quality of life. Am J Transplant 2019;19:2825–32. doi:10.1111/ajt.15350
      OpenUrl
    9. ↵
      1. Kehlet H,
      2. Dahl JB
      . Anaesthesia, surgery, and challenges in postoperative recovery. Lancet 2003;362:1921–8. doi:10.1016/S0140-6736(03)14966-5
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. Perkins FM,
      2. Kehlet H
      . Chronic pain as an outcome of surgery. A review of predictive factors. Anesthesiology 2000;93:1123–33. doi:10.1097/00000542-200010000-00038
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      1. Widmar M,
      2. Keskin M,
      3. Strombom PD, et al
      . Evaluating the validity of the Clavien-Dindo classification in colectomy studies: a 90-day cost of care analysis. Dis Colon Rectum 2021;64:1426–34. doi:10.1097/DCR.0000000000001966
      OpenUrl
    12. ↵
      1. Li Z,
      2. Bai B,
      3. Ji G, et al
      . Relationship between Clavien-Dindo classification and long-term survival outcomes after curative resection for gastric cancer: a propensity score-matched analysis. Int J Surg 2018;60:67–73. doi:10.1016/j.ijsu.2018.10.044
      OpenUrlCrossRefPubMed
    13. ↵
      1. Wang D,
      2. Zhang J,
      3. Bai Z, et al
      . Associations of postoperative complications assessed by Clavien-Dindo classification and comprehensive complication index with long-term overall survival in elderly patients after radical CRC resection. Clin Interv Aging 2020;15:1939–49. doi:10.2147/CIA.S271969
      OpenUrl
    14. ↵
      1. Dumitra S,
      2. O’Leary M,
      3. Raoof M, et al
      . The comprehensive complication index: a new measure of the burden of complications after hyperthermic intraperitoneal chemotherapy. Ann Surg Oncol 2018;25:688–93. doi:10.1245/s10434-017-6157-z
      OpenUrl
    15. ↵
      1. Buchholz V,
      2. Hazard R,
      3. Yin Z, et al
      . The impact of intraoperative and postoperative fluid balance on complications for transthoracic esophagectomy: a retrospective analysis. BMC Res Notes 2023;16:315. doi:10.1186/s13104-023-06574-x
    16. ↵
      1. DeOliveira ML,
      2. Winter JM,
      3. Schafer M, et al
      . Assessment of complications after pancreatic surgery: a novel grading system applied to 633 patients undergoing pancreaticoduodenectomy. Ann Surg 2006;244:931–7. doi:10.1097/01.sla.0000246856.03918.9a
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      1. Clavien PA,
      2. Barkun J,
      3. de Oliveira ML, et al
      . The Clavien-Dindo classification of surgical complications: five-year experience. Ann Surg 2009;250:187–96. doi:10.1097/SLA.0b013e3181b13ca2
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      1. Bolliger M,
      2. Kroehnert JA,
      3. Molineus F, et al
      . Experiences with the standardized classification of surgical complications (Clavien-Dindo) in general surgery patients. Eur Surg 2018;50:256–61. doi:10.1007/s10353-018-0551-z
      OpenUrlCrossRefPubMed
    19. ↵
      1. Mitropoulos D,
      2. Artibani W,
      3. Biyani CS, et al
      . Validation of the Clavien-Dindo grading system in Urology by the European Association of Urology guidelines ad hoc panel. Eur Urol Focus 2018;4:608–13. doi:10.1016/j.euf.2017.02.014
      OpenUrl
    20. ↵
      1. Dindo D,
      2. Demartines N,
      3. Clavien PA
      . Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240:205–13. doi:10.1097/01.sla.0000133083.54934.ae
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      1. Muthén LK,
      2. Muthén BO
      . Mplus user’s guide. 7th edn. 2015.
    22. ↵
      1. Cohen J
      , ed. Statistical power analysis for the behavioral sciences. 2nd edn. 1988.
    23. ↵
      1. Enders CK
      . Applied missing data analysis. New York London, 2010.
    24. ↵
      1. Lee T,
      2. Shi D
      . A comparison of full information maximum likelihood and multiple imputation in structural equation modeling with missing data. Psychol Methods 2021;26:466–85. doi:10.1037/met0000381
      OpenUrlCrossRef
    25. ↵
      1. Jakobsen JC,
      2. Gluud C,
      3. Wetterslev J, et al
      . When and how should multiple imputation be used for handling missing data in randomised clinical trials - a practical guide with Flowcharts. BMC Med Res Methodol 2017;17:162. doi:10.1186/s12874-017-0442-1
    26. ↵
      1. Box GEP,
      2. Tidwell PW
      . Transformation of independent variables. Technometrics 1962;4:531–50. doi:10.1080/00401706.1962.10490038
      OpenUrlCrossRefWeb of Science
    27. ↵
      1. Field AP
      . Discovering statistics using IBM SPSS statistics. 2018.
    28. ↵
      1. Gaskell AL,
      2. Sleigh JW
      . An introduction to causal diagrams for Anesthesiology research. Anesthesiology 2020;132:951–67. doi:10.1097/ALN.0000000000003193
      OpenUrlCrossRefPubMed
    29. ↵
      1. Neuss H,
      2. Reetz C,
      3. Raue W, et al
      . Postoperative surgical complications after radical Axillary lymph node dissection in Melanoma disease result in increased pain. Int Surg 2010;95:166–71.
      OpenUrlPubMed
    30. ↵
      1. Zajączkowska R,
      2. Leppert W,
      3. Mika J, et al
      . Perioperative immunosuppression and risk of cancer progression: the impact of opioids on pain management. Pain Res Manag 2018;2018:9293704. doi:10.1155/2018/9293704
    31. ↵
      1. Tennant F
      . The physiologic effects of pain on the endocrine system. Pain Ther 2013;2:75–86. doi:10.1007/s40122-013-0015-x
      OpenUrl
    32. ↵
      1. Beilin B,
      2. Shavit Y,
      3. Trabekin E, et al
      . The effects of postoperative pain management on immune response to surgery. Anesth Analg 2003;97:822–7. doi:10.1213/01.ANE.0000078586.82810.3B
      OpenUrlCrossRefPubMedWeb of Science
    33. ↵
      1. Jänig W
      . Sympathetic nervous system and inflammation: a conceptual view. Auton Neurosci 2014;182:4–14. doi:10.1016/j.autneu.2014.01.004
      OpenUrlCrossRefPubMed
    34. ↵
      1. Aston-Jones G,
      2. Waterhouse B
      . Locus coeruleus: from global projection system to adaptive regulation of behavior. Brain Res 2016;1645:75–8. doi:10.1016/j.brainres.2016.03.001
      OpenUrlCrossRefPubMed
    35. ↵
      1. Benarroch EE
      . Locus coeruleus. Cell Tissue Res 2018;373:221–32. doi:10.1007/s00441-017-2649-1
      OpenUrlCrossRef
    36. ↵
      1. Zaslansky R,
      2. Rothaug J,
      3. Chapman CR, et al
      . PAIN OUT: the making of an international acute pain Registry. Eur J Pain 2015;19:490–502. doi:10.1002/ejp.571
      OpenUrlPubMed
    37. ↵
      1. van Helden EV,
      2. van Uitert A,
      3. Albers KI, et al
      . Chronic postsurgical pain after minimally invasive adrenalectomy: prevalence and impact on quality of life. BMC Anesthesiol 2022;22:153. doi:10.1186/s12871-022-01696-4
    38. ↵
      1. Fan X,
      2. Wang S,
      3. Pan K, et al
      . Selective COX-2 inhibitor is beneficial in suppressing chronic postsurgical pain in esophageal cancer patients and may prolong patient survival. Oncol Res Treat 2023;46:503–10. doi:10.1159/000535183
      OpenUrl
    39. ↵
      1. Wang X,
      2. Li T
      . Postoperative pain pathophysiology and treatment strategies after CRS + HIPEC for peritoneal cancer. World J Surg Oncol 2020;18:62. doi:10.1186/s12957-020-01842-7
    40. ↵
      1. Gerbershagen HJ,
      2. Aduckathil S,
      3. van Wijck AJM, et al
      . Pain intensity on the first day after surgery: a prospective cohort study comparing 179 surgical procedures. Anesthesiology 2013;118:934–44. doi:10.1097/ALN.0b013e31828866b3
      OpenUrlCrossRefPubMedWeb of Science
    41. ↵
      1. Sacks GD,
      2. Lawson EH,
      3. Dawes AJ, et al
      . Relationship between hospital performance on a patient satisfaction survey and surgical quality. JAMA Surg 2015;150:858–64. doi:10.1001/jamasurg.2015.1108
      OpenUrl

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Footnotes

    • Contributors EvH: conception and design of the study, data acquisition, analysis and interpretation of data, drafting and revising the manuscript critically for important intellectual content and final approval of the version to be published. JK: data acquisition, analysis and interpretation of data, drafting and revising the manuscript critically for important intellectual content and final approval of the version to be published. AV: statistical analysis, interpretation of data, revising the manuscript critically for important intellectual content and final approval of the version to be published. AdB: conception and design of the study, analysis and interpretation of data, revising the manuscript critically for important intellectual content, and final approval of the version to be published. PdR: data collection and management, revising the manuscript critically for important intellectual content and final approval of the version to be published. CR: revising the manuscript critically for important intellectual content and final approval of the version to be published. JdW: conception and design of the study, interpretation of data, revising the manuscript critically for important intellectual content and final approval of the version to be published. KvL: final approval of the version to be published. GJS and CK: supervision, revising the manuscript critically for important intellectual content and final approval of the version to be published. MW: supervision, conception and design of the study, analysis and interpretation of data, drafting and revising the manuscript critically for important intellectual content and final approval of the version to be published. EH and MW

      are responsible for the overall content as the guarantor.

    • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

    • Competing interests None declared.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.