The “Weekday Effect”—Does It Impact Esophageal Cancer Surgery Outcomes?

Increased 30-day mortality rates have been reported in patients undergoing elective surgery later compared with earlier in the week. However, these reports have been conflicting for esophageal surgery. We conducted a study to assess the differences in outcomes of patients undergoing surgery for esophageal cancer earlier in the week (Tuesday) versus later (Friday). This retrospective analysis of a prospectively maintained database included patients with esophageal cancer who underwent esophageal resection in a tertiary cancer center between 1 January 2005 and 31 December 2017. We compared patients operated on Tuesdays versus Fridays. The primary outcome was a composite of major morbidity (defined as Clavien-Dindo grade 3 or more) and/or mortality. Secondary outcomes included duration of post-operative ventilation, and length of ICU and hospital stay. Among 1300 patients included, 733 were operated on a Tuesday and 567 on a Friday. Patient and surgery characteristics were similar in the two groups. The primary outcome (composite of major morbidity and mortality) was 23.6% in the Tuesday group versus 26.3% in the Friday group. Mortality was similar in the two groups (6.0%). Multivariable logistic regression analysis showed that the day of surgery was not a predictor of major morbidity or mortality. In patients undergoing esophagectomy at tertiary care high volume cancer center, there was no difference in major morbidity and mortality whether the surgery was performed early in the week (Tuesday) or closer to the weekend (Friday).


Background
Esophageal cancer is one of the common cancers worldwide, as well as in India.It is also a leading cause of cancerrelated mortality [1,2].Although a multimodality approach has become the standard of care for the past few decades, surgery is still the mainstay of treatment for localized and loco-regionally advanced esophageal cancer wherever feasible [3][4][5].Surgery for esophageal cancer is complex, extensive, and associated with a learning curve [6].Morbidity and mortality are more frequent after esophagectomy than most other oncological operations, and have been shown to be associated with hospital as well as surgeon volume [7].The immediate post-operative period after esophagectomy is considered crucial and an important determinant of postoperative outcomes as well as prognosis [8].
A meta-analysis of approximately 100 studies [9] evaluated outcomes of patients hospitalized during the weekend, and found that they had significantly higher mortality-the so called weekend effect.This effect has also been studied in elective as well as emergency surgeries [10][11][12].As an extension to the weekend effect, several studies, including a meta-analysis of cohort studies looking at outcomes after both elective and emergent surgery, have shown increased post-operative mortality after elective surgery performed during the week but closer to the weekend, termed the "weekday effect."[13,14] The mechanisms suggested are surgeon and surgery team fatigue as the week progresses, along with shortage of staff, access to imaging, and other investigations.Most of these studies have been retrospective audits [14], whereas some were large nationwide [15] and international cohorts [16].
Large studies [17][18][19] have shown conflicting results of the impact of esophagectomy performed earlier in the week compared to that performed later in the week.Lagergren and colleagues evaluated 1748 patients undergoing esophagectomy from 1987 to 2010 with follow-up until 2014 and found that there was no association between weekday of surgery (Monday or Tuesday versus Wednesday to Friday) and reoperation or mortality within 30 days of surgery [17].However, they found a 13% increase on all-cause 5-year mortality when surgery was performed later in the week compared to Monday and Tuesday [18].Another study analyzed the Netherlands Cancer Registry Database from 2005 to 2013 and did not find any impact of weekday on 30-day mortality, the total number of resected lymph nodes, R0 resection rates, and overall long-term survival [19].
The aim of our study was to assess the difference in the outcomes of patients undergoing esophagectomy on a Tuesday versus on a Friday at a large volume tertiary cancer center.We hypothesized that major morbidity and mortality would be higher in patients undergoing esophagectomy later in the week, as compared to those operated earlier in the week.
The primary objective of the study was to compare groups with respect to a composite outcome of major morbidity (as assessed by Clavien-Dindo grade 3 or more) and in-hospital mortality while the secondary objectives included comparison of post-operative length of ICU and hospital stay between the groups.

Methods
We performed a retrospective analysis of a prospectively maintained database of patients who underwent esophagectomy between 1 January 2005 and 31 December 2017 in a large tertiary cancer center.Approval was obtained from the institutional ethics committee along with a waiver of consent.We included patients who underwent esophageal resection with curative intent (both upfront and after neoadjuvant therapy), on Tuesdays and Fridays during this period.We excluded patients who were found to have unresectable tumors and those operated on Tuesdays where the following Wednesday was a hospital holiday.
From the database, we extracted data on the following: 1. Patient characteristics (age, sex, pre-operative American Society of Anesthesiologists status, pre-chemotherapy, extent of disease (TNM stage), neo-adjuvant therapy received.2. Surgery characteristics: type and approach of surgery, surgeon grade (trainee/consultant), duration of surgery, intra-operative blood loss.3. Outcomes: the primary outcome was a composite of major morbidity (defined as Clavien-Dindo grade 3 or more) and/or mortality (Including in-hopital and 30-day mortality, whichever occurred later).Secondary outcomes included duration of post-operative ventilation, and length of ICU and hospital stay.
This study was conducted at a tertiary cancer center, one of the largest institutions in the country.During the period of the study, the thoracic surgical team consisted of three faculty, with 5-20 years of experience in treating esophageal cancer along with 7-8 residents, fellows, and trainees at any given point in time.Of these, at least 3 residents were available for any emergencies during the night.We also had three specialized thoracic anesthetists, a chest physiotherapist, 3 pulmonologists, and an ICU team fully equipped to manage patients undergoing major cancer surgeries.The radiology and intervention radiology departments were equipped to handle emergency CT scans and procedures respectively throughout the week, having residents on call during the night, with the ICU and radiology/intervention radiology staffing during the night for emergency procedures, on the weekend being the same as that during the week.
There were no specific factors that would entail scheduling a patient on a Tuesday versus a Friday in terms of infrastructure, equipment, or personnel required to carry out major surgery safely.There was no variation between the availability of the thoracic surgery and anesthesia faculty as well as residents between surgeries performed on a Tuesday versus surgeries performed on a Friday.
All patients with a confirmed diagnosis of esophageal cancer and fit to undergo curative resection are appropriately staged and discussed in a multidisciplinary joint clinic.All patients with evidence of disease more than T2N0 on clinicoradiological staging were referred for neoadjuvant therapy and re-assessed for surgery at 2-3 weeks after completion of the same.Patients were scheduled for surgery between 4 and 6 weeks after the last dose of neoadjuvant therapy.The thoracic surgery operative days were Tuesdays, Thursdays, and Fridays.Most patients underwent a transthoracic esophagectomy, either via open or via minimally invasive approach.
In accordance with the ERAS protocol followed by our unit, an attempt was made to extubate all patients on table.Patients were observed in the surgical ICU overnight for monitoring and if stable, they were shifted to the surgical ward on the next day.All patients were started on enteral feeding on postoperative day 1 through a nasojejunal tube, and orals were encouraged from post-operative days 4-5, after documenting lack of vocal cord palsy on indirect laryngeal examination.Patients who developed complications such as recurrent laryngeal nerve palsy, pulmonary complications, and anastomotic leaks had modifications in their post-operative course according to standard guidelines.
Data was entered into a statistical software (SPSS 20.0) for analysis.We compared outcomes between the two groups using the chi-square test for categorical variables, and the unpaired t-test or Mann-Whitney test for continuous variables.A multivariable logistic regression model was used to study the association between day of surgery and primary outcome, adjusted for age, gender, ASA grade, neoadjuvant treatment, type of surgery, and extent of lymphadenectomy.

Results
Between 1 Jan 2005 and 31 Dec 2017, a total of 2269 cases of esophagectomy were performed, with the main thoracic surgical operative days being Tuesday, Thursday, and Friday.Of these, we excluded 801 cases operated on other days (not Tuesday/Friday) and 39 cases operated on Tuesdays which were followed by holidays on Wednesday.
Therefore, 1429 cases operated on a Tuesday and a Friday were eligible for inclusion.Of these, 118 were inoperable and 11 had missing details.A total of 1300 were available for analysis with 733 surgeries performed on a Tuesday (56%) and 567 on a Friday (44%) as summarized in Fig. 1.
The mean age and sex ratio of the patients was similar in both groups.The groups were well matched with respect to pre-operative clinic-radiological stage, ASA physical status classification, and neoadjuvant treatment as seen in Table 1.
No major differences were seen in the operative characteristics such as surgical approach, grade of the operating surgeon, operative time, and blood loss.Most of the patients (approximately 80% in both groups) underwent transthoracic esophagectomy.
The American Society of Anesthesiology (ASA) grading was not different for patients undergoing surgery on a Tuesday versus on a Friday, with an ASA grade of II seen in 32% and ASA grade III in 0.7% of patients undergoing surgery on a Tuesday versus 28% grade II and  being operated on a Friday.The cardiac, pulmonary, renal, and hepatic risks were also similar among the two groups.
There was no difference between the two groups in the primary composite outcome or any of the secondary outcomes (Table 2) of major morbidity.Among patients undergoing surgery on a Tuesday, 172 patients (23.6%) experienced the primary composite endpoint of major complications or mortality compared to 149 patients (26.3%) undergoing surgery on a Friday.Overall mortality was 6% in both groups.The length of post-operative ventilator support and length of ICU and hospital stay were similar in the two groups as seen in Table 2. Pulmonary complications between the two groups were not dissimilar, 23% of the patients operated on a Friday needed prolonged ventilation (defined as needing postoperative ventilatory support for more than 24 hours) versus 21.6% of those operated on a Tuesday (p = 0.42).Also, 16.5% of those undergoing surgery on a Friday were re-ventilated versus 13.8% of those operated on a Tuesday (p = 0.17).Approximately 31% of the patients in both groups had any pulmonary complications, with only 1% in both groups suffering from a life-threatening pulmonary complication.Recurrent laryngeal nerve palsy was seen in 35.3% of the Friday group versus 32.05% of the Tuesday group (p = 0.31).
Anastomotic leak was seen in 21.4% of the patients undergoing surgery on a Friday versus 16% of those operated on a Tuesday (p = 0.85).Minor anastomotic leaks were seen in 8.3% in the Friday group versus 6.9% in the Tuesday group whereas major anastomotic leaks (requiring radiological intervention, delayed orals, and prolonged hospital stay) in 9.3% and 5.8% respectively.Anastomotic leaks requiring surgical intervention were 2.9% in both groups.Arrhythmias, cardiac and renal complications, chyle leaks, and wound infections were alike in both groups.
Of the 567 patients operated on a Friday, 56 patients had major complications (Clavien-Dindo grade 3 and above) versus 70 out of 733 operated on a Tuesday.There was no significant difference in mortality due to major complications between the two groups, 57% of the patients died due to major complication after being operated on a Friday versus 60% of those operated on a Tuesday (p = 0.086) as seen in Table 3.
On multivariate regression analysis, age, neoadjuvant treatment, and transthoracic approach were independent predictors of post-operative major morbidity and/or death but day of surgery, extent of lymphadenectomy, and ASA grade did not show any association (Table 4).

Discussion
In our study, 24.8% (322 out of 1300) patients undergoing planned esophagectomy for cancer at a high-volume tertiary care cancer center had major morbidity and/or mortality, with no difference in outcomes regardless of whether the surgery was performed early in the week or closer to the weekend.Similarly, there were no differences in any of the other outcomes studied including ICU and hospital stay.Esophagectomy is the only curative treatment for operable esophageal cancer but is associated with significant morbidity and mortality [20][21][22] with an estimated 50 to 60% of esophagectomy patients developing a complication, and 30-and 90-day mortality rates ranging widely between 2 and 23% [23].In surgical specialties, it has been hypothesized that procedures performed on the weekend have higher complication rates than those performed on weekdays [24].A large part of this data comes from emergency surgeries and retrospective studies with ambivalent results [25][26][27][28][29].
Globally, elective surgeries are rarely performed on the weekend.However, patients operated for elective procedures at the end of the week are likely to need most of their immediate post-operative care over the weekend when resources are possibly strained.Inadequate staffing may not only be quantitative, but also qualitative, in terms of more senior and experienced medical staff being on leave during the weekend.There may be a lower nursing-to-patient ratio on weekends, with fewer nurses and adjunct technical staff in high-intensity-care areas [14].Also, general practitioners and referring physicians may tend to refer the sick and elderly patients to a hospital setting closer to the weekend rather than caring for them in the community setting [31].Therefore, it has been hypothesized that because of this "weekday effect," patients undergoing elective surgeries towards the end of the week are likely to fare worse than those operated early in the week [24].
But the evidence pointing towards this effect is unclear [9,14].Aylin et al. in the UK investigated 27,582 deaths (within 30 days) after 4,133,346 inpatient admissions for elective operating room procedures; overall crude mortality rate was 6.7 per 1000.The adjusted odds of death were 44% and 82% higher, respectively, if the procedures were carried out on a Friday (OR = 1.44, 95% CI 1.39-1.50)or a weekend (OR = 1.82, 95% CI 1.71-1.94)compared with a Monday [14].Pauls et al. [9] retrospectively evaluated 402,899 Canadian elective surgeries performed over a period of 10 years (2002-2012) and found that although less experienced surgeons operated closer to a Friday, there was no difference in 30-day mortality irrespective of the day of the week when the surgery was performed.
Cancer surgeries are major procedures with patients needing intensive care in the post-operative period and, therefore, are likely to be maximally impacted by the weekday effect, if it exists.Njjolstad et al. [32] found that surgical treatment of endometrial cancers later in the week was associated with worse outcomes.In contrast, Huijts et al. [33] and Frostberg et al. [34] found no evidence of weekday effect in patients undergoing elective surgery for colon or rectal cancer.Similarly, Visser et al. [35] found no difference in immediate or long-term outcomes after gastrectomy done earlier versus later in the week.Their only significant finding was that lymph node yield was significantly lower, later in the week.
Very few previous studies [17][18][19] have looked at the impact of day of surgery on outcome after esophagectomy.In an analysis of a large retrospective database of patients who underwent esophagectomy, Visser et al. [19] found no difference in post-operative complications, lymph node yield, or overall survival between patients operated on different weekdays.A paper by Lagergren showed that long-term outcomes were worse for esophagectomy patients operated later in the week than earlier; this association was strongest for patients with early-stage disease and absent for those with advanced disease [18].However, another paper by the same group showed that day of the week did not influence post-operative complications or 30-day mortality [17].
There are several limitations with the available data on the weekday effect in general.First, there are no randomized trials evaluating the effect of weekday or weekend on short-or long-term hospital outcomes; however, we feel that a randomized trial is highly unlikely.All the available data consists of analyses of large observational datasets, with their inherent biases despite statistical adjustments for known confounders.
Second, many of the studies which have demonstrated the presence of a weekend or weekday effect suffer from the problem of multiple comparisons increasing the likelihood of spurious association [36].Third, in many of the studies, there is the possibility of inaccurate coding of diagnoses at admission.In a review examining the weekend effect in patients with stroke, Li et al. [37] showed that the finding of a weekend effect was seen in studies that used only administrative coding.In contrast, no weekend effect was observed in studies where clinical adjudication of stroke diagnosis was used.Finally, the findings in many of the studies have no scientific basis.For example, patients with advanced esophageal cancer are likely to have more complex surgery with a higher probability of complications.Contrary to the findings in Lagergren's study, such patients should be more affected by the weekday effect than those with early disease.
A systematic review [38] showed that at weekends, patients who are admitted to hospital are more severely ill with differences in care pathways before and after admission [30].This is more likely to reflect in semi-urgent and emergency indications for hospital admission.This could provide a possible explanation for the absence of the weekday effect in our study because we focused on a site-specific elective surgery where all procedures were performed at a high-volume oncology center by a small group of trained surgeons, and standard well-established protocols were followed peri-operatively, irrespective of the day of the week.
Our study has several strengths-this was a prospectively maintained database with all the coding performed by a clinician and not by administrative staff.The database was analyzed retrospectively; therefore, the person performing the coding was not aware of the study hypothesis, thus minimizing bias.To avoid problems with multiple comparisons, we used a limited number of endpoints.To the best of our knowledge, this is the first large series to study the weekday effect in esophageal cancer surgery outside the Netherlands.Except for the fact that our study was not a randomized trial (which may be challenging to conduct), it is a robust study that adds to the limited data on the weekday effect in esophagectomy.

Conclusion
In summary, the findings of our study show no association between day of surgery and post-operative outcomes in patients undergoing esophagectomy at a high-volume tertiary care cancer center and prove that major surgery is safe and feasible throughout the week.

Table 1
Patient, tumor, and surgery characteristics Data is expressed as absolute numbers with percentages in parentheses for categorical data and mean with standard deviation in parentheses for numerical data Data is expressed as absolute numbers with percentages in parentheses for categorical data and mean with standard deviation in parentheses for numerical data *Skewed data for which medians and interquartile ranges (IQR) have been reportedTuesday (n = 733) Friday (n =

Table 4
Predictors of major morbidity and/or mortality