Abstract
Background
Textbook outcome (TO) is a multidimensional measure reflecting the ideal outcome after surgery. As a benchmarking tool, it provides an objective overview of quality of care. Uniform definitions of TO in hepato-pancreato-biliary (HPB) surgery are missing. This study aimed to provide a definition of TO in HPB surgery and identify obstacles and predictors for achieving it.
Methods
A systematic literature search was conducted using PubMed, Embase, and Cochrane Database according to PRISMA guidelines. Studies published between 1993 and 2021 were retrieved. After selection, two independent reviewers extracted descriptive statistics and derived summary estimates of the occurrence of TO criteria and obstacles for achieving TO using co-occurrence maps.
Results
Overall, 30 studies were included. TO rates ranged between 16–69 per cent. Commonly chosen co-occurring criteria to define TO included ‘no prolonged length of stay (LOS)’, ‘no complications’, ‘no readmission’, and ‘no deaths’. Major obstacles for achieving TO in HPB surgery were prolonged LOS, complications, and readmission. On multivariable analysis, TO predicted better overall and disease-free survival in patients with cancer. Achievement of TO was more likely in dedicated centres and associated with procedural and structural indicators, including high case-mix index and surgical volume.
Conclusion
TO is a useful quality measure to benchmark surgical outcome. Future definitions of TO in HPB surgery should include ‘no prolonged LOS’, ‘no complications’, ‘no readmission’, and ‘no deaths’.
Textbook outcome (TO) reflects the ideal outcome after surgery and is a useful quality measure to benchmark surgical outcome. We are the first to propose a uniform definition of TO in hepato-pancreato-biliary surgery, including ‘no prolonged length of stay’, ‘no complications’, ‘no readmission’, and ‘no mortality’.
Introduction
Hepato-pancreato-biliary (HPB) surgery frequently involves complex procedures, notably those operations undertaken for malignant disease. Many procedures are prone to complications that influence postoperative course and can have a detrimental effect on outcome. Advances in surgical technique and perioperative care have made surgery safer, leading to increased proclivity to consider patients with more advanced disease or co-morbidities as candidates for surgery1,2. In light of these changes, reliable measures that assess patient-centred outcome and overall quality of surgical care are indispensable. Ideally, these measures will help guide treatment decisions and ensure that treatment meets appropriate standards.
Textbook outcome (TO) is a composite measure, originally described in colorectal cancer surgery, that aims to reflects the ideal surgical outcome in a single indicator3. TO is achieved when all prespecified parameters are fulfilled according to an all-or-none principle. In this regard, TO can provide a global picture and an overall reflection of surgical quality and care. In turn, identification of parameters that have the greatest influence on TO achievement can help to target single issues and make specific changes accordingly to improve healthcare quality. In this respect, TO has been suggested as a new benchmarking tool to measure internal surgical quality and provide quality-of-care information to patients. Indeed, patients prefer summary measures regarding quality-of-care information4,5.
A uniform definition of TO in HPB surgery is presently missing6–9. Ideally, included parameters should be disease and surgery specific.
This study aimed to systematically review literature to investigate existing definitions of TO in HPB surgery and provide the basis for a uniform definition of this composite measure. The primary objective was to assess parameters used for defining TO. Secondary objectives were TO achievement rate; identifications of factors influencing TO achievement; relation between TO and minimally invasive surgery, type of resection (minor versus major), survival, hospital performance (procedural volume, case-mix index (CMI)), type of hospital, hospital designation (including Magnet status), and socioeconomic factors.
Methods
Search strategy
A systematic literature search was conducted using PubMed, Embase, and Cochrane Database according to the PRISMA guidelines10 by two independent investigators. All studies published in English until December 2021 were potentially eligible for inclusion. The search terms were (textbook outcome) OR (textbook AND outcome*). The reference list of each article was searched for further relevant literature. Duplicate articles, editorials, and conference abstracts were excluded. The articles were screened and filtered by title and abstract. A full-text assessment and review of the remaining studies was conducted. Data from the included articles were extracted independently by the two reviewers using double-data extraction. Inconsistencies were resolved by consensus. In the case of disagreement, a third reviewer was consulted so that consensus was reached (Fig. 1).
Fig. 1.
Flow chart of literature selection strategy
Eligibility criteria
All studies reporting TO in patients 18 years or older who underwent either hepatic, biliary, or pancreatic surgery or a combination of these for any condition (benign or malignant) were considered eligible for inclusion. Multivisceral resections including other compartments (for example rectal resection) were excluded.
Data extraction and analysis
The following data were collected: author details, year of publication, country, recruitment interval, study population, sample size, diagnosis, procedure, definition of TO with all included components (including the percentage of the respective components reached), occurrence of TO, independent predictors of TO, any data with regard to hospital status and performance (procedural volume, CMI11, teaching status, and Magnet designation12), expenditure, social vulnerability, and racial diversity.
Data visualization
Co-occurrence matrix
Co-occurrence maps were computed to visualize how many times each TO criterion was considered. Co-occurrence is defined as an above-chance frequency of occurrence of two terms/criteria13. These maps show TO criteria along the x and y axis, with matrix values (and tile colours) indicating the amount of co-occurrence. Co-occurrence maps were generated first including all studies on hepatobiliary surgery for both benign and malignant disease (Fig. 2a) and second, focusing on only subgroup analyses of pancreatic cancer (Fig. 2c) and only hepatobiliary malignancies (Fig. 2d). Summary graphs were also provided (Fig. 2b). Summary variables included ‘no prolonged length of stay (LOS)’ in either the 50th or 75th percentile, ‘no readmission’, and ‘no deaths’ at 30 days, 90 days, in-hospital, and any time, ‘no complications’ either overall or three or more, and ‘no bile leak’ grade B/C, or any as defined by the International Study Group for Liver Surgery14.
Fig. 2.
Definition of textbook outcome
Co-occurrence maps visualizing how many times each TO criterion was examined in the context of all other criteria across studies. TO criteria are shown along the x and y axis, with matrix values (and tile colours) indicating the amount of co-occurrence. a All HPB studies, similar TO criteria were grouped into summary variables: ‘no prolonged LOS’ (summarizing: no prolonged LOS in either the 50th or 75th percentile), ‘no readmission’ (30 days, 90 days, any), ‘no deaths’ (30 days, 90 days, in-hospital, any), ‘no complications’ (three or more, any), and ‘no bile leak’ (grade B/C, any), b All HPB studies, similar TO criteria are all shown individually, c Studies focusing on pancreatic cancer only. d Studies focusing on hepatobiliary malignancies only. HPB, hepato-pancreato-biliary; TO, textbook outcome; CTx, chemotherapy; LOS, length of stay; POPF, postoperative pancreatic fistula; PPH, post-pancreatectomy haemorrhage.
TO and specific parameter achievement rates
Rates of patients that achieved each TO criterion were derived from the original publication or, where not available, computed by the authors of this study. In the presence of subgroups, averages were weighted by the number of participants within each subgroup.
Multivariable contributors
To help readers visualize the contribution of each parameter to TO achievement, negative (indicated in blue) and positive (indicated in red) contributors were displayed using word clouds. The size of each word/term reflects the strength of its contribution to TO. Where identical terms were present across multiple studies, their OR values were averaged. Studies with missing values were excluded. To indicate the relevance of each contributor to achieving TO in the context of a particular diagnosis, diagnostic labels in parentheses following terms (where applicable) were added.
Results
Literature search
A total of 650 studies were identified after de-duplication. Of these, 14 conference abstracts were excluded. Screening by title led to the exclusion of 576 studies, 22 studies were excluded after abstract review, leaving 38 studies that underwent full-text assessment, with 30 meeting eligibility criteria (Fig. 1). The eight studies excluded did not reported the relevant outcomes.
Overall, 10 studies used administrative data from Medicare Inpatient and Outpatient Standard Analytic Files with in-part overlapping cohorts12,15–23. As they were focused on different quality metrics, they were all included in a purely descriptive manner to give the broadest overview on the current knowledge on TO.
Study characteristics
The characteristics of the included studies are shown in Table 1. All studies retrieved information from prospectively maintained databases. Study intervals ranged from 1993 to 2020. Two studies started before 2000, 12 studies before 2010, and the remaining 16 studies from 2010 onwards. More than half of the included studies were from the USA (16), six were international, and five were from France. The remaining studies were from UK (one), The Netherlands (one), Spain, and France (one binational study). There were 28 multi-institutional studies. The sample size of patients per study ranged from 78 to 42 551. Ten studies had a sample size lower than 1000, six between 1000–10 000, and 14 more than 10 000.
Table 1.
Characteristics of the included studies
| Author | Diagnosis | Procedure | n | Recruitment interval | Overall TO rate (%) | TO rate (%) subgroups |
|---|---|---|---|---|---|---|
| Pancreatic surgery | ||||||
| Aquina et al.24 | Malignancy | Pancreatic surgery | 42 551 | 2006–2017 | 25 | |
| Beane et al.25 | Any indication | Pancreatic surgery | 24 168 | 2013–2017 | NA | PD: 55.4 DP: 55.8 |
| Diaz et al.18 | Any indication | Pancreatic surgery | 24 298 | 2013–2017 | 43.3 | Low diversity area: 40 Average diversity area: 43 High diversity area: 49 |
| Heidsma et al.26 | PNET | Pancreatic surgery | 821 | 2000–2016 | 49.3 | PD: 32.5 DP: 56.7 enucleation: 52 |
| Kulshrestha et al.27 | PDAC | PD | 16 602 | 2006–2015 | 21.5 | high volume: 26.2 moderate volume: 18.5 low volume: 17.3 very low volume: 12.2 |
| Lof et al.28 | Any indication | PD (only open) | 375 | 2009–2017 | NA | ERAS: 56.4 Non-ERAS: 44 |
| Mehta et al.21 | Malignancy | Pancreatic surgery | 26 268 | 2016–2017 | NA | Trifactor status: 49.7 Non-trifactor status: 42.9 |
| Merath et al.29 | Any indication | Pancreatic surgery | 4853 | 2013–2015 | NA | |
| Sweigert et al.6 | PDAC | PD | 18 608 | 2006–2016 | 16.8 | Over time, in 2007: 12.9 Over time, in 2015: 19.5 |
| Sweigert et al.30 | PDAC | PD | 12 854 | 2010–2015 | NA | Laparoscopic: 24.7 Open: 23.5 |
| Van Roessel et al8 | Any indication | Pancreatic surgery | 3341 | 2014–2017 | 60.3 | PD: 58.3 DP: 67.4 |
| Liver surgery | ||||||
| Azoulay et al.31 | HCC (CSPH ≥ 10mmHg) | Liver surgery | 79 | 1999–2019 | 34 | |
| Brustia et al.32 | ICC | Liver surgery | 855 | 2000–2018 | 15.8 | Laparoscopic: 30.8 Open: 12.7 |
| Görgec et al.9 | Any indication | Liver surgery | 8188 | 2011–2019 | 69.1 | Laparoscopic: 74.8 Open: 61.9 |
| Hobeika et al.33 | HCC | Liver surgery | 425 | 2010–2018 | 32.9 | Laparoscopic: 38.7 Open: 24.2 |
| Hobeika et al.34 | Any indication | Liver surgery (only laparoscopic) | 1343 | 2000–2017 | NA | left lateral sectionectomy: 43.7 right hepatectomy: 23.8 |
| Hobeika et al.35 | ICC | Liver surgery | 548 | 2000–2017 | 22.1 | Laparoscopic: 28.3 Open: 20.2 |
| Merath et al.36 | ICC | Liver surgery | 687 | 1993–2015 | 25.5 | Eastern countries: 25.7 Western countries: 66.7 |
| Nassar et al.37 | Malignancy | Liver surgery (only laparoscopic) | 463 | 2009–2019 | NA | Difficulty level A: 39.1 Difficulty level B: 36.8 Difficulty level C: 30.8 |
| Paro et al.23 | Any indication | Liver surgery | 13 898 | 2013–2017 | 53.2 | MetS: 46.8 No MetS: 54.6 |
| Tsilimigras et al.38 | HCC | Liver surgery | 605 | 2000–2015 | 62.3 | |
| Tsilimigras et al.39 | HCC, ICC | Liver surgery | 1829 | 2005–2017 | 62 | HCC: 68 ICC: 55 |
| Yoshino et al.40 | Any indication | Liver surgery | 78 | 2000–2020 | 38 | |
| Hepatopancreatic surgery | ||||||
| Azap et al.17 | Any indication | Hepatopancreatic surgery | 32 142 | 2013–2017 | 51.2 | Resection pancreas: 46.6 Resection liver: 57.6 Low SVI: 52.8 Intermediate SVI: 51.5 High SVI: 49.8 |
| Hyer et al.11 | Any indication | Hepatopancreatic surgery | 31 452 | 2013–2017 | 50.7 | Low CMI: 45.7 Average CMI: 47.5 High CMI: 53 |
| Mehta et al.20 | Malignancy | Hepatopancreatic surgery | 35 352 | 2013–2015 | NA | Resection pancreas: 37.8 Resection liver: 38.5 |
| Mehta et al.19 | Malignancy | Hepatopancreatic surgery | 8035 | 2013–2015 | 44.3 | Minor teaching status: 40 Major teaching status: 45.4 |
| Mehta et al.22 | Malignancy | Hepatopancreatic surgery | 21 234 | 2013–2017 | 45.6 | |
| Mehta et al.12 | Malignancy | Hepatopancreatic surgery | 10 997 | 2015–2017 | 45.6 | Magnet status: 46 Non-Magnet status: 45 |
| Merath et al.15 | Any indication | Hepatopancreatic surgery | 13 467 | 2013–2015 | 44 | Minor resection pancreas: 47.8 Major resection pancreas: 24.7 Minor resection liver: 46.8 Major resection liver: 33.3 |
CMI, case-mix index; CSPH, clinically significant portal hypertension; DP, distal pancreatectomy; ERAS, enhanced recovery after surgery; HCC, hepatocellular carcinoma; ICC, intrahepatic cholangiocellular carcinoma; MetS, metabolic syndrome; NA, not available; PD, pancreatoduodenectomy; PNET, pancreatic neuroendocrine tumour; SVI, social-vulnerability index.
Patient characteristics
Most studies were focused on patients receiving HPB surgery for a malignant indication (n = 18). Among those, six included only patients with pancreatic ductal adenocarcinoma (PDAC), eight focused on liver cancer (three hepatocellular carcinoma (HCC), three intrahepatic cholangiocarcinoma (ICC), one HCC and ICC, and one primary and secondary malignancy), and four looked at HPB malignancies as a whole.
Definition of textbook outcome
Parameters used to define TO are shown in Fig. 2. The most frequently used and co-occurring measure were ‘no prolonged LOS’, ‘no readmission’, ‘no deaths’, and ‘no complications’ (Fig. 2a). Multiple definition existed for these parameters. ‘No prolonged LOS higher or equal to the 75th percentile’, ‘no readmission within 90 days’, ‘no deaths within 90 days’, and ‘no complications’ were the most used (Fig. 2b). Figure 2c,d shows the results for subgroup analysis with regard to PDAC (Fig. 2c) and hepatobiliary malignancies (Fig. 2d).
Achievement of textbook outcome
The rate of patients achieving TO ranged between 15.8–69.1 per cent, 16.8–60.3 per cent, and 15.8–69.1 per cent in all candidates, pancreatic surgery candidates (including benign and malignant disease), and liver surgery candidates (including benign and malignant disease) respectively. Median TO rate was 38 per cent. Patients with malignant disease presented lower TO rates than the median and patients with benign disease. Patients with benign disease had a TO rate higher than the median. (Table 1).
Major obstacles in achieving textbook outcome
Figure 3 gives an overview of the parameters chosen to define TO within each study, including the percentage of patients reaching each component. Accordingly, the major obstacles for achieving TO across all studies were ‘prolonged LOS’, ‘complications’, and ‘readmission’.
Fig. 3.
Components of textbook outcome and percentages reached
Overview of criteria chosen to define TO within each study with percentages of patients reaching each component (studies with missing data are not included). The colour gradient highlights that the major obstacles for achieving TO across all studies were ‘prolonged LOS’, ‘complications’, and ‘readmission’. LOS, length of stay; read, readmission; mort, deaths; comp, complication; transf, transfusion; LAE, lymphadenectomy; adj CTx, adjuvant chemotherapy; POPF, postoperative pancreatic fistula; PPH, post-pancreatectomy haemorrhage; TO, textbook outcome.
Textbook outcome and operative approach
Seven studies investigated the TO rate with regard to the operating approach. All studies found better TO rates in patients undergoing a laparoscopic procedure compared with an open procedure. Two of these studies included only patients with ICC32,35 and three studies compared the laparoscopic and open approach in patients with HCC, portal hypertension and cirrhosis, and patients with any liver disease9,31,33. In both studies focusing on HCC, the laparoscopic approach proved to be a predictor of TO in multivariable analysis with Ors of 2.81 (95 per cent c.i. 1.29 to 6.12) and 5.6 (95 per cent c.i. 1.7 to 18.2)31,33. One study investigating patients who underwent HPB surgery for any indication reported that a laparoscopic approach was an independent predictor of TO achievement with an OR of 1.52 (95 per cent c.i. 1.34 to 1.73)15. One study investigating TO with respect to operating approach among patients with PDAC failed to identify a difference in TO rates between minimally invasive and open pancreaticoduodenectomy30.
Textbook outcome, type, and extent of operation
TO ranged between 32.5–58.3 per cent, and 55.8–67.4 per cent after pancreaticoduodenectomy and distal pancreatectomy respectively8,25,26. PDAC (OR 1.36, 95 per cent c.i. 1.14 to 1.63) and a dilated pancreatic duct (more than 3 mm) (OR 2.22, 95 per cent c.i. 2.05 to 3.57) were identified as independent prognostic factors of TO achievement after pancreatoduodenectomy (OR 1 or higher) in one study8.
Among hepatobiliary procedures, previous liver resection, ICC, gallbladder carcinoma, increasing tumour size (more than 3 cm), minor resection of posterior/superior segments, non-anatomical resection, (anatomical) major resection, biliary reconstruction, and major vascular invasion were all associated with decreased likelihood of achieving TO (OR 1 or less)9,15,23,33,39.
Textbook outcome, hospital status, and performance
Case-mix index
Hospital CMI was strongly associated with the probability of achieving TO in one study. TO rate was 45.7 per cent and 53 per cent in facilities with low and high CMI respectively. A low as well as an average CMI were significantly associated with lower odds of achieving TO (OR 0.78, 95 per cent c.i. 0.69 to 0.87 and OR 0.82, 95 per cent c.i. 0.76 to 0.88 respectively) when compared with high CMI. Even after adjusting for HPB surgical volume at the hospital level, CMI had a beneficial effect on the odds of achieving TO11.
Surgical volume
Twelve studies evaluated the relationship between TO and surgical volume. Seven studies reported a significant positive correlation, whereas five did not (Table 2).
Table 2.
Textbook outcome and surgical volume
| Author | Diagnosis | Specific volume parameters assessed | OR (95% c.i.) | P |
|---|---|---|---|---|
| Multivariate analysis | ||||
| Kulshrestha et al.27 | Pancreatic cancer | High volume (versus very low) | 2.39 (2.02–2.85) | P < 0.001 |
| Mehta et al.12 | Hepatopancreatic malignancies | Magnet centre: Leapfrog volume compliant (versus noncompliant), Non-Magnet centre: Leapfrog volume compliant (versus noncompliant) |
1.24 (1.06–1.44), 1.18 (1.11–1.26) |
P < 0.001 |
| Hobeika et al.34 | Any liver disease | In Laparoscopic right hepatectomies: ≥ 35 liver resections/year, In laparoscopic left lateral sectionectomies: ≥ 25 liver resections/year |
2.55 (1.34–5.63), 2.45 (1.65–3.69) |
P < 0.001 |
| Mehta et al.19 | Hepatopancreatic malignancies | High volume (versus low) | 1.30 (1.15–1.47), 1.19 (1.03–1.38) |
P < 0.001 |
| Sweigert et al.6 | Pancreatic cancer | Low volume (<20 PD/year) (versus high) | 0.49 (0.38–0.64) | P < 0.001 |
| Merath et al.29 | Any pancreatic disease | Leapfrog volume compliant (versus noncompliant) | 1.28 (1.09–1.50) | P < 0.001 |
| Univariate analysis | ||||
| Azap et al.17 | Any hepatopancreatic disease | High volume (versus low) | P < 0.001 | |
| Merath et al.15 | Any hepatopancreatic disease | Not further specified | P > 0.050 | |
| Tsilimigras et al.38 | HCC | Not further specified | P > 0.050 | |
| Van Roessel et al.8 | Any pancreatic disease | Not further specified | P > 0.050 | |
| Görgec et al.9 | Any liver disease | Not further specified | P > 0.050 | |
| Merath et al.36 | ICC | Not further specified | P > 0.050 | |
HCC, hepatocellular carcinoma; ICC, intrahepatic cholangiocellular carcinoma.
Hospital designation
Four studies investigated the relationship between hospital designation and TO. Two studies showed that hospitals meeting the quality trifactor (Leapfrog minimum volume standards, Hospital Safety Grades, and Magnet recognition) had higher odds of TO (OR 1.37, 95 per cent c.i. 1.21 to 1.55 and OR 1.28, 95 per cent c.i. 1.03 to 1.59) and lower odds of complications, prolonged LOS, and deaths21,29.
One study suggested that dedicated cancer centres provide higher-value surgical care for patients with HPB malignancies. Patients treated in dedicated cancer centres, even when presenting with a higher co-morbidity burden, had higher odds of achieving TO (pancreatic surgery OR 1.71, 95 per cent c.i. 1.50 to 1.95; liver surgery OR 1.36, 95 per cent c.i. 1.15 to 1.62) compared with those treated in other hospitals22.
Socioeconomic and segregation factors
Patients with a high social-vulnerability index (SVI) were less likely to achieve TO after HPB surgery (pancreatic surgery OR 0.89, 95 per cent c.i. 0.82 to 0.97; and liver surgery OR 0.89, 95 per cent c.i. 0.80 to 0.98) in one study. A high SVI was an independent predictor of complications and 90-day deaths after both pancreatic and liver surgery as well as prolonged LOS in liver surgery17. Socioeconomic factors independently associated with achievement of TO were lower patient education, government insurance, low county-level racial diversity, and Black race6,18,24,27.
Expenditure
Four studies investigated the relationship between expenditure (insurance payments) and TO. All reported that payments for patients achieving TO were markedly lower (about $10 000 less) than payments for patients who did not achieve TO15,17,19,22,23.
Textbook outcome and survival
Eleven studies assessed the impact of TO on survival. While two studies did not report better survival rates when TO was achieved32,35, the remaining nine studies did 6,24,26,27,30,31,33,38,39 (Table S1).
Tsilimigras et al. reported that among patients with HCC who achieved TO, 5-year overall survival (OS) was higher (69.6 per cent versus 56.9 per cent) compared with patients who did not. TO achievement was associated with 40 per cent decreased risk of death with respect to all-cause deaths (HR 0.60, 95 per cent c.i. 0.42 to 0.85)38. Another study reported that TO was independently associated with 26 per cent and 37 per cent decreased hazards of death among ICC (HR 0.74, 95 per cent c.i. 0.56 to 0.97) and patients with HCC (HR 0.63, 95 per cent c.i. 0.46 to 0.85) respectively39. Similarly, in a study by Sweigert et al. on patients with PDAC, TO achievement was associated with better OS (median 27 versus 19.9 months) and lower risk of long-term deaths (HR 0.73, 95 per cent c.i. 0.70 to 0.77)6. Three studies on PDAC concluded that TO was associated with better OS (TO median, 26.7 months versus no TO median, 21.1 months)30, a 4.8-month increase in median survival at any facility27, and that all components of TO (in this study: R0, adequate lymphadenectomy, no prolonged LOS, no readmission, and start of adjuvant chemotherapy within 12 weeks after surgery) were independently associated with the 5-year OS24. Ultimately, two studies investigating the impact of TO on disease-free survival (DFS) in patients with HCC and pancreatic neuroendocrine tumours reported that TO was independently associated with DFS (HR 0.34, 95 per cent c.i. 0.19 to 0.60 and HR 0.49, 95 per cent c.i. 0.32 to 0.75 respectively)26,33.
Textbook outcome, patient characteristics, and perioperative factors
Several studies investigated patient characteristics and perioperative factors and their association to TO achievement (Fig. 4).
Fig. 4.
Multivariate contributors to textbook outcome
For ease of interpretation, negative (OR less than 1), indicated in blue, a and positive (OR higher than 1, indicated in red, b contributors are visualized using word clouds. The size of each word/term reflects the strength of its (negative/positive) contribution to TO. Specifically, weak and strong contributions are represented by small and large font. To indicate the relevance of each contributor to achieving TO in the context of a particular diagnosis, we added diagnostic labels in parentheses following terms (where applicable). L, hepatic surgery; HP, hepatobiliary surgery; PD, pancreatoduodenectomy; TO, textbook outcome; ALBI, albumin-bilirubin; ICC, intrahepatic cholangiocarcinoma; MELD, Model for End-Stage Liver Disease; CMI, case-mix index; SVI, social-vulnerability index; CTx, chemotherapy; BCLC, Barcelona Clinic Liver Cancer; PDAC, pancreatic ductal adenocarcinoma.
Decreased odds of TO (OR 1 or lower)
Parameters associated with decreased odds of achieving TO in multivariate analysis included Charlson–Deyo co-morbidity index 3 or higher, older age (age more than 70 years), and neoadjuvant chemotherapy6,24,27 for patients with PDAC, and an ASA score 3 or higher in patients with any pancreatic disease receiving pancreatoduodenectomy8. Age more than 65 years, an ASA score more than 2 (or 3 or higher), the Model for End-Stage Liver Disease score, and an albumin-bilirubin grade 2/3 were independent predictors of TO in patients with hepatic malignancies33,39. Metabolic syndrome, ASA score of 3 or higher, and previous hepatic surgery were negatively associated with TO in patients undergoing liver surgery for any liver disease9,23.
More than one previous hospital admission was associated with decreased odds of TO (odds decreasing by the increasing number of previous stays) in patients undergoing HPB surgery for any indication15.
Increased odds of TO (OR more than 1)
Female sex was an independent predictor of TO in patients undergoing HPB surgery both for cancer or any indication8,15,27,28. Omission of neoadjuvant chemotherapy in patients undergoing distal pancreatectomy, and a dilated pancreatic duct (more than 3 mm) as well as diagnosis of PDAC in patients receiving pancreatoduodenectomy were associated with higher likelihood of achieving TO8. A Barcelona clinic liver cancer stage 0 (versus B/C) and an albumin-bilirubin grade 1 (versus 2/3) were independent predictors of TO in patients with HCC38. In patients with ICC, age 60 years or younger, no preoperative jaundice, and omission of neoadjuvant chemotherapy increased the likelihood of achieving TO36.
Discussion
In this systematic review the median rate of TO achievement was 38 per cent, ranging widely from 15 to 70 per cent. Major obstacles in achieving TO were ‘complications’, ‘prolonged LOS’, and ‘readmission’. While identification of these parameters might be of great benefit to guide improvement initiatives, it is important to note that they are themselves influenced by patient and procedural factors. LOS and readmission not only depend on surgical quality but also on cultural norms and healthcare policies as payment schemes and access to long-term rehabilitation facilities41.
Many factors influenced TO achievement. Laparoscopic approach, less-extended surgery (distal pancreatectomy versus pancreaticoduodenectomy and minor versus major liver resection), high CMI and surgical volume, and higher socioeconomic status all influenced TO positively. Age, co-morbidity status, and neoadjuvant chemotherapy were also reported as independent predictors of TO after multivariable analysis in some studies8,9,23,33,39. Indeed, the retrospective fashion of all these studies severely limits the reliability of this data.
Limitations of this review included heterogeneity of definitions of TO that impeded fair comparisons of TO rates between institutions so that meta-analysis was inappropriate. The retrospective nature of the studies and use of administrative data sets from large databases might have led to selection or secular bias and lack of relevant clinical variables and perioperative parameters12, 15–23.
TO definition per se has some flaws. TO parameter selections must be conducted carefully, and likelihood of achievement should not influence this decision as it would lead to biased high TO rates as in a self-fulfilling prophecy (for example choosing parameters that positively influence laparoscopy in a centre that performs mainly laparoscopic surgery). Aiming to reach high TO rate should not lead to operations being offered only to the ‘best’ candidates and avoiding those at higher risk. To avoid these biases, TO should be used in addition to other quality metrics such as CMI11,12,20–22,29. TO does not include a patient perspective. As patient-reported and clinically defined outcomes may well differ, future studies should further investigate this relationship and incorporate patient-reported outcomes in TO definitions7.
TO is a promising multidimensional measure reflecting the ideal outcome after surgery but is still heterogeneously defined. A current definition of TO in HPB surgery should include the terms ‘no prolonged LOS’, ‘no complications’, ‘no readmission’, and ‘no deaths’. Under proper risk adjustment, TO in addition to other quality metrics, might become a useful tool to aid clinical decision-making, provide information on surgical quality for the patient, and assist preoperative patient selection.
Supplementary Material
Acknowledgements
We give special thanks to C.M. Pretzsch for her assistance in the development of figures.
Contributor Information
Elise Pretzsch, Department of General, Visceral, and Transplant Surgery, Ludwig-Maximilians-University Munich, Munich, Germany.
Dionysios Koliogiannis, Department of General, Visceral, and Transplant Surgery, Ludwig-Maximilians-University Munich, Munich, Germany.
Jan Gustav D’Haese, Department of General, Visceral, and Transplant Surgery, Ludwig-Maximilians-University Munich, Munich, Germany.
Matthias Ilmer, Department of General, Visceral, and Transplant Surgery, Ludwig-Maximilians-University Munich, Munich, Germany.
Markus Otto Guba, Department of General, Visceral, and Transplant Surgery, Ludwig-Maximilians-University Munich, Munich, Germany.
Martin Konrad Angele, Department of General, Visceral, and Transplant Surgery, Ludwig-Maximilians-University Munich, Munich, Germany.
Jens Werner, Department of General, Visceral, and Transplant Surgery, Ludwig-Maximilians-University Munich, Munich, Germany.
Hanno Niess, Department of General, Visceral, and Transplant Surgery, Ludwig-Maximilians-University Munich, Munich, Germany.
Funding
The authors have no funding to declare.
Disclosure
The authors declare no conflict of interest.
Supplementary material
Supplementary material is available at BJS Open online.
Data availability
The data that support the findings of this study are available on request from the corresponding author.
References
- 1. Perera SK, Jacob S, Wilson BEet al. Global demand for cancer surgery and an estimate of the optimal surgical and anaesthesia workforce between 2018 and 2040: a population-based modelling study. Lancet Oncol 2021;22:182–189 [DOI] [PubMed] [Google Scholar]
- 2. Cloyd J, Mizuno T, Kawaguchi Yet al. Comprehensive complication Index validates improved outcomes over time despite increased complexity in 3707 consecutive hepatectomy. Ann Surg 2020;271:724–731 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Kolfschoten NE, Kievit J, Gooiker GAet al. Focusing on desired outcomes of care after colon cancer resections; hospital variations in “textbook outcome”. Eur J Surg Oncol 2013;39:156–163 [DOI] [PubMed] [Google Scholar]
- 4. De Mheen PJM-V, Dijs-Elsinga J, Otten Wet al. The relative importance of quality of care information when choosing a hospital for surgical treatment: a hospital choice experiment. Med Decis Mak 2011;31:816–827 [DOI] [PubMed] [Google Scholar]
- 5. Dijs-Elsinga J, Otten W, Versluijs MMet al. Choosing a hospital for surgery: the importance of information on quality of care. Med Decis Making 2010;30:544–555 [DOI] [PubMed] [Google Scholar]
- 6. Sweigert PJ, Eguia E, Baker MSet al. Assessment of textbook oncologic outcomes following pancreaticoduodenectomy for pancreatic adenocarcinoma. J Surg Oncol 2020;121:936–944 [DOI] [PubMed] [Google Scholar]
- 7. Wiseman JT, Sarna A, Wills CEet al. Patient perspectives on defining textbook outcomes following major abdominal surgery. J Gastrointest Surg 2022;26:197–205 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Van Roessel S, Mackay TM, Van Dieren Set al. Textbook outcome: nationwide analysis of a novel quality measure in pancreatic surgery. Ann Surg 2020;271:155–162 [DOI] [PubMed] [Google Scholar]
- 9. Görgec B, Benedetti Cacciaguerra A, Lanari Jet al. Assessment of textbook outcome in laparoscopic and open liver surgery. JAMA Surg 2021;156:e212064. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Shamseer L, Moher D, Clarke Met al. Preferred reporting Items for Systematic Review and Meta-Analysis protocols (PRISMA-p) 2015: elaboration and explanation. BMJ 2015;349:g7647–g7647 [DOI] [PubMed] [Google Scholar]
- 11. Hyer JM, Tsilimigras DI, Diaz A, Mirdad RS, Pawlik TM. A higher hospital case mix index increases the odds of achieving a textbook outcome after hepatopancreatic surgery in the Medicare population. Surgery (United States) 2021;170:1525–1531 [DOI] [PubMed] [Google Scholar]
- 12. Mehta R, Tsilimigras DI, Pawlik TM. Assessment of magnet status and textbook outcomes among Medicare beneficiaries undergoing hepato-pancreatic surgery for cancer. J Surg Oncol 2021;124:334–342 [DOI] [PubMed] [Google Scholar]
- 13. Nanni L, Brahnam S, Ghidoni S, Menegatti E, Barrier T. Different approaches for extracting information from the co-occurrence matrix. PloS ONE 2013;8:e83554. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Brooke-Smith M, Figueras J, Ullah Set al. Prospective evaluation of the international study group for liver surgery definition of bile leak after a liver resection and the role of routine operative drainage: an international multicentre study. HPB (Oxford) 2015;17:46–51 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Merath K, Chen Q, Bagante Fet al. Textbook outcomes among Medicare patients undergoing hepatopancreatic surgery. Ann Surg 2020;271:1116–1123 [DOI] [PubMed] [Google Scholar]
- 16. Hyer JM, Beane JD, Spolverato Get al. Trends in textbook outcomes over time: are optimal outcomes following complex gastrointestinal surgery for cancer increasing? J Gastrointest Surg 2022;26:50–59 [DOI] [PubMed] [Google Scholar]
- 17. Azap RA, Paredes AZ, Diaz A, Hyer JM, Pawlik TM. The association of neighborhood social vulnerability with surgical textbook outcomes among patients undergoing hepatopancreatic surgery. Surgery (United States) 2020;168:868–875 [DOI] [PubMed] [Google Scholar]
- 18. Diaz A, Dalmacy D, Herbert C, Mirdad RS, Hyer JM, Pawlik TM. Association of county-level racial diversity and likelihood of a textbook outcome following pancreas surgery. Ann Surg Oncol 2021;26:8076–8084 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Mehta R, Paredes AZ, Tsilimigras DIet al. Influence of hospital teaching status on the chance to achieve a textbook outcome after hepatopancreatic surgery for cancer among Medicare beneficiaries. Surgery (United States) 2020;168:92–100 [DOI] [PubMed] [Google Scholar]
- 20. Mehta R, Tsilimigras DI, Paredes AZet al. Comparing textbook outcomes among patients undergoing surgery for cancer at US News & World Report ranked hospitals. J Surg Oncol 2020;121:927–935 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Mehta R, Tsilimigras DI, Paredes Aet al. Assessment of hospital quality and safety standards among Medicare beneficiaries undergoing surgery for cancer. Surgery (United States) 2021;169:573–579 [DOI] [PubMed] [Google Scholar]
- 22. Mehta R, Tsilimigras DI, Paredes AZet al. Dedicated cancer centers are more likely to achieve a textbook outcome following hepatopancreatic surgery. Ann Surg Oncol 2020;27:1889–1897 [DOI] [PubMed] [Google Scholar]
- 23. Paro A, Tsilimigras D, Dalmacy D, Mirdad R, Hyer J, Pawlik T. Impact of metabolic syndrome on postoperative outcomes among Medicare beneficiaries undergoing hepatectomy. J Gastrointest Cancer 2021;25:2545–2552 [DOI] [PubMed] [Google Scholar]
- 24. Aquina CT, Hamad A, Becerra AZet al. Is textbook oncologic outcome a valid hospital-quality metric after high-risk surgical oncology procedures? Ann Surg Oncol 2021;28:8028–8045 [DOI] [PubMed] [Google Scholar]
- 25. Beane JD, Borrebach JD, Zureikat AH, Kilbane EM, Thompson VM, Pitt HA. Optimal pancreatic surgery: are we making progress in North America? Ann Surg 2021;274:e355–e363 [DOI] [PubMed] [Google Scholar]
- 26. Heidsma CM, Hyer M, Tsilimigras DIet al. Incidence and impact of textbook outcome among patients undergoing resection of pancreatic neuroendocrine tumors: results of the US neuroendocrine Tumor Study Group. J Surg Oncol 2020;121:1201–1208 [DOI] [PubMed] [Google Scholar]
- 27. Kulshrestha S, Sweigert P, Tonelli Cet al. Textbook oncologic outcome in pancreaticoduodenectomy: do regionalization efforts make sense? J Surg Oncol 2021;125:414–424 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Lof S, Benedetti Cacciaguerra A, Aljarrah Ret al. Implementation of enhanced recovery after surgery for pancreatoduodenectomy increases the proportion of patients achieving textbook outcome: a retrospective cohort study. Pancreatology 2020;20:976–983 [DOI] [PubMed] [Google Scholar]
- 29. Merath K, Mehta R, Tsilimigras DIet al. Quality of care among Medicare patients undergoing pancreatic surgery: safety grade, magnet recognition, and Leapfrog minimum volume standards—which quality benchmark matters? J Gastrointest Surg 2021;25:269–277 [DOI] [PubMed] [Google Scholar]
- 30. Sweigert PJ, Wang X, Eguia Eet al. Does minimally invasive pancreaticoduodenectomy increase the chance of a textbook oncologic outcome? Surgery (United States) 2021;170:880–888 [DOI] [PubMed] [Google Scholar]
- 31. Azoulay D, Ramos E, Casellas-Robert Met al. Liver resection for hepatocellular carcinoma in patients with clinically significant portal hypertension. JHEP Rep 2021;3:100190. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Brustia R, Laurent A, Goumard Cet al. Laparoscopic versus open liver resection for intrahepatic cholangiocarcinoma: report of an international multicenter cohort study with propensity score matching. Surgery (United States) 2021;171:1290–1302 [DOI] [PubMed] [Google Scholar]
- 33. Hobeika C, Nault JC, Barbier Let al. Influence of surgical approach and quality of resection on the probability of cure for early-stage HCC occurring in cirrhosis. JHEP Rep 2020;2:100153. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Hobeika C, Fuks D, Cauchy Fet al. Benchmark performance of laparoscopic left lateral sectionectomy and right hepatectomy in expert centers. J Hepatol 2020;73:1100–1108 [DOI] [PubMed] [Google Scholar]
- 35. Hobeika C, Cauchy F, Fuks Det al. Laparoscopic versus open resection of intrahepatic cholangiocarcinoma: nationwide analysis. Br J Surg 2021;108:419–426 [DOI] [PubMed] [Google Scholar]
- 36. Merath K, Chen Q, Bagante Fet al. A multi-institutional international analysis of textbook outcomes among patients undergoing curative-intent resection of intrahepatic cholangiocarcinoma. JAMA Surg 2019;154:1–9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Nassar A, Tribillon E, Marchese Uet al. Feasibility and outcomes of multiple simultaneous laparoscopic liver resections. Surg Endosc 2021;36:2466–2472 [DOI] [PubMed] [Google Scholar]
- 38. Tsilimigras DI, Mehta R, Merath Ket al. Hospital variation in textbook outcomes following curative-intent resection of hepatocellular carcinoma: an international multi-institutional analysis. HPB (Oxford) 2020;22:1305–1313 [DOI] [PubMed] [Google Scholar]
- 39. Tsilimigras DI, Sahara K, Moris Det al. Assessing textbook outcomes following liver surgery for primary liver cancer over a 12-year time period at major hepatobiliary centers. Ann Surg Oncol 2020;27:3318–3327 [DOI] [PubMed] [Google Scholar]
- 40. Yoshino K, Hamzaoui Y, Yoh Tet al. Liver resection for octogenarians in a French center: prolonged hepatic pedicle occlusion and male sex increase major complications. Langenbeck’s Arch Surg 2021;406:1543–1552 [DOI] [PubMed] [Google Scholar]
- 41. Tiessen J, Kambara H, Sakai T, Kato K, Yamauchi K, Mcmillan C. What causes international variations in length of stay: a comparative analysis for two inpatient conditions in Japanese and Canadian hospitals. Heal Serv Manag Res 2013;26:86–94 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author.