Risk factors and economic burden of postoperative anastomotic leakage related events in patients who underwent surgeries for colorectal cancer.

BACKGROUND: Nationwide research about the clinical and economic burden caused by anastomotic leakage (AL) has not been published yet in Korea. This study assessed the AL rate and quantified the economic burden using the nationwide database.
METHODS: This real world evidence study used health claims data provided by the Korean Health Insurance Review and Assessment Service (HIRA, which showed that 156,545 patients underwent anterior resection (AR), low anterior resection (LAR), or ultra-low anterior resection (uLAR) for colorectal cancer (CRC) between January 1, 2007 and January 31, 2020. The incidence of AL was identified using a composite operational definition, a composite of imaging study, antibacterial drug use, reoperation, or image-guided percutaneous drainage. Total hospital costs and length of stay (LOS) were evaluated in patients with AL versus those without AL during index hospitalization and within 30 days after the surgery.
RESULTS: Among 120,245 patients who met the eligibility criteria, 7,194 (5.98%) patients had AL within 30 days after surgery. Male gender, comorbidities (diabetes, metastatic disease, ischemic heart disease, ischemic stroke), protective ostomy, and multiple linear stapler use, blood transfusion, and urinary tract injury were associated with the higher odds of AL. Older age, rectosigmoid junction cancer, AR, LAR, and laparoscopic approach were related with the reduced odds of AL. Patients with AL incurred higher costs for index hospitalization compared to those without AL (8,991 vs. 7,153 USD; p<0.0001). Patients with AL also required longer LOS (16.78 vs. 14.22 days; p<0.0001) and readmissions (20.83 vs. 13.93 days; p<0.0001).
CONCLUSION: Among patients requiring resection for CRC, the occurrence of AL was associated with significantly increased costs and LOS. Preventing AL could not only produce superior clinical outcomes, but also reduce the economic burden for patients and payers.

Introduction

Colorectal cancer (CRC) was the fourth most common cancer following lung, breast, and prostate cancer, and was the second leading cause of cancer related deaths around the world in 2018 [1]. According to the annual report of cancer statistics in Korea in 2017, CRC is the second most common cancer type, with 28,111 new cases and the third leading cause of cancer-related deaths behind lung and liver cancer [2]. Mutiple surgical approaches, including hemicolectomy (right or left), anterior resection, and low anterior resection, are commonly indicated for patients with CRC [3].

Complications of colorectal resection can be divided into intraoperative and postoperative ones. Intraoperative complications include bleeding, bowel injuries, ureteral lesions, and bladder injuries. The most frequent postoperative complications include surgical site infection, anastomotic leakage (AL), intraabdominal abscess, ileus, and bleeding [4]. Among these, AL is considered to be the most detrimental in terms of the impact on morbidity, mortality and quality of life [5-8]. AL leads to anastomotic stricture and impaired colorectal function including reduced neorectal capacity, evacuation problems, fecal urgency, and incontinence [9, 10]. More seriously, AL is associated with an increase in local recurrence and lower long-term survival [11-14]. Alongside these complications and the potential need for reoperation, AL management consumes extensive healthcare resources and expenses [8, 15–17].

The rate of AL has been reported to range from 5% to 19% globally in colorectal/coloanal surgery depending upon perioperative factors such as anatomical site, surgical method, operators’ experience, and patient’s clinical characteristics [6, 18]. Although a multicenter study reported an AL rate of 6.3% after laparoscopic rectal cancer surgery in Korea, the AL rate at each hospital showed wide variation ranging from 2.0% to 10.3% [19]. To identify the generalizable economic burden caused by AL, a nationwide population study is necessary but has not been carried out yet in Korea.

This retrospective, nationwide claims analysis was conducted to identify clinical and economic outcomes among Korean patients with cancer who had undergone anterior resection (AR), low anterior resection (LAR), or ultra-low anterior resection (uLAR) with a manual circular stapler as the current standard of care. This study sought to assess an association between such factors as patients, procedures, and providers and the incidence of AL, and its economic burden in terms of length of stay (LOS), readmissions, and total costs.

Methods

Data sources

Under the public and single-payer system, health claims data have been generated in the archives of the Health Insurance Review and Assessment Service (HIRA) in the process of reimbursing healthcare providers, covering all Korean citizens. As these reimbursements are predominantly based on the fee-for-service system, claims data contain comprehensive information about treatments, pharmaceuticals, procedures, and diagnoses for almost 50 million beneficiaries. This study was approved by the Public Institutional Bioethics Committee designated by the Ministry of Health & Welfare (MOHW) (No. P01-202007-21-024) including the waiver of informed consent as it was a retrospective study based on a de-identified claims database.

Study population

The HIRA database identified 156,545 patients who underwent AR, LAR, or uLAR (procedure codes: Q2921, QA921, Q2922, QA922, Q2928, QA928) between January 1, 2007 and January 31, 2020, of whom only adult patients (≥19 years) who had undergone surgery as the principal treatment for CRC (ICD-10 code: C18, C19, C20) using manual circular staplers (device code: B1022XXX) were included. After excluding patients who had prior colorectal surgery within a 1-year lookback period, the analysis dataset included a total of 120,245 patients. The “index surgery” was defined as the first colorectal resection for each patient, and the “index hospitalization” as admission for the index surgery.

Clinical outcomes

The following variables were identified as the indicators of surgical complications within 30 days after the index surgery: AL, blood transfusion, urinary tract injury, pulmonary embolism, acute renal failure.

Since there was no specific diagnosis code for AL, the following procedures were required for AL occurrence during in-hospital care to meet operational AL definitions: (1) Imaging study including computed tomography scans (2) Administration of antibacterial drugs (more than 7 consecutive days after the surgery) (3) Abdominal reoperation and (4) image-guided percutaneous drainage. AL occurrence is defined by the combination of the above four items. Combinations to meet the definition of AL are any including both (1) and (2), any including (3), or any including (4). Among these combinations to define AL, those that include (3) or (4) were designated as additional intervention cases (AIC), because reoperation and percutaneous drainage are potentially more definitive treatments for AL compared to imaging and antibiotics.

Blood transfusion was detected using the procedure code. Urinary tract injury, pulmonary embolism, and acute renal failure were all also identified through the ICD-10 codes.

Economic outcomes

The economic outcomes identified were as follows: LOS for the index hospitalization, LOS for all-cause readmissions within 30 days, total costs for the index hospitalization, total costs for during the period from the index surgery to 30 days post-operation, and total costs for readmissions. To comprehensively understand the economic burden on society, total costs included both out-of-pocket expenditures for patients and costs covered by payer.

Covariates

The following covariates of interest were included: demographics (age, gender), patient characteristics (primary diagnosis for the index surgery, Charlson Comorbidity Index (CCI), other comorbidities (diabetes, chronic obstructive pulmonary disease (COPD), metastatic disease, ischemic heart disease, ischemic stroke), pre-operation radiation therapy, procedure characteristics (surgical procedure, surgical approach, protective ostomy, multiple circular stapler use, multiple linear stapler use), provider characteristics (number of beds) and others (blood transfusion, urinary tract injury).

Statistical analysis

All study variables were analyzed descriptively. Continuous variables were presented as mean, median, and standard deviation, while categorical variables were expressed as frequencies and proportions.

The overall incidence of each complication (AL, additional intervention cases, blood transfusion, urinary tract injury, pulmonary embolism, acute renal failure) was identified, and the incidence of AL was further summarized by surgical procedure (AR, LAR, or uLAR), primary diagnosis (C18, C19, C20) and surgical approach (open or laparoscopic).

To figure out the factors associated with AL, all covariates considered in this study were compared according to the incidence of AL. Student’s t-test or Wilcoxon’s rank sum test was applied for comparison of continuous variables while a chi-squared test or Fisher’s exact test was conducted for categorical variables as appropriate after testing assumptions such as normality. The univariable and multivariable logistic regression models were fitted to assess the risk factors associated with AL so that the odds ratio of each risk factor could be calculated. The multivariable model included factors that showed the statistical significance based on univariable analysis. LOS and costs were estimated during a 30-day period after the index surgery. Costs were summarized by surgical procedure and approach and compared with the Kruskal-Wallis (K-W) test. When the costs difference was significant by the K-W test, then the Dwass, Steel, Critchlow-Fligner method was applied for post hoc comparison.

All analyses were conducted using SAS (version 9.2, SAS Institutes, Cary, NC, USA) and R (version 4.0.3, The R Foundation for statistical computing, Vienna, Austria). A p-value of 0.05 or less was considered to indicate statistically significant difference.

Ethical considerations

The study was reviewed by the Public Institutional Review Board designated by the Ministry of Health and Welfare and determined to be exempt from IRB approval (Review number: P01-202007-21-024). As the study involves no more than minimal risk to patients, IRB approved a request to waive informed consent under the Bioethics and Safety Act of Korea.

Results

Incidence of anastomotic leakage and additional interventional cases based on the operational definition

Among 156,545 patients who underwent AR, LAR, or uLAR from January 2007 to January 2020, a total of 120,245 patients met the eligibility criteria (Fig 1) for this study. Among these patients, 33.97% had imaging studies, 7.54% received antibacterial drugs for more than 7 consecutive days, 0.36% had reoperations, and 6.99% had image-guided percutaneous drainage (Table 1). 5.98% of patients satisfied the operational definition of AL having events corresponding with the defined combination within 30 days after surgery. When applying a stricter definition that requires reoperation or image-guided percutaneous drainage, 4.56% of patients fulfilled the definition of additional interventional cases (AIC).

Fig 1

Pateint selection flow chart.

Table 1

Anastomotic leakage and other clinical outcomes.

	Total
	(N = 120,245)
Item 1 (Imaging Study)	40,851	(33.97%)
Item 2 (Administration of antibacterial drug)	9,068	(7.54%)
Item 3 (Abdominal reoperation)	427	(0.36%)
Item 4 (Image guided percutaneous drainage)	8,406	(6.99%)
Anastomotic leakage (within 30 days)
No	113,051	(94.02%)
Yes	7,194	(5.98%)
Combination to meet the definition of AL
Combination 1 (all two items of 1 and 2 without 3 or 4)	1,711	(23.78%)†
Combination 2 (any combination including 3 without 4)	40	(0.56%)†
Combination 3 (any combination including 4 without 3)	5,291	(73.55%)†
Combination 4 (any combination including 3 and 4)	152	(2.11%)†
Additional intervention cases (within 30 days)
No	114,762	(95.44%)
Yes	5,483	(4.56%)
Combination to meet the definition of AIC
Combination 2 (any combination including 3 without 4)	40	(0.73%)††
Combination 3 (any combination including 4 without 3)	5,291	(96.50%)††
Combination 4 (any combination including 3 and 4)	152	(2.77%)††

†Denominator = Number of patients with anastomotic leakage (7,194)

††Denominator = Number of patients with additional intervention cases (5,483).

Demographics and perioperative clinical characteristics

Patient demographics and perioperative characteristics are displayed in Table 2. Their mean age was 64.09 (standard deviation [SD] ± 11.45) years, and there was no statistically significant difference between patients with and without AL. The proportion of males was significantly higher in patients with AL (66.68% vs. 62.41%, p<0.0001). More patients with AL had LAR or uLAR than those without AL (LAR: 61.44% vs. 53.57% and uLAR: 2.93% and 1.61%, p<0.0001). Patients with AL had more open surgery (32.50% vs. 25.47%, p<0.0001), protective ostomy (27.15% vs. 15.83%, p<0.0001), multiple linear stapler use (57.83% vs. 48.94%, p<0.0001), multiple circular stapler use (1.08% vs. 0.76%, p = 0.0027), and blood transfusion (32.44% vs. 17.00%, p<0.0001) than those without AL. The following comorbidities were more frequently observed in patients with AL than those without AL; diabetes (24.48% vs. 20.62%, p<0.0001), COPD (7.17% vs. 6.33%, p = 0.0046), metastatic disease (12.65% vs. 7.41%, p<0.0001), ischemic heart disease (10.19% vs. 8.22%, p<0.0001), and ischemic stroke (5.35% vs. 3.97%, p<0.0001). The CCI score in the AL group was significantly higher than that in patients without AL (mean ± SD, 3.75 ± 2.90 vs. 3.18 ± 2.43, p<0.0001).

Table 2

Demographics and perioperative clinical characteristics.

	With AL		Without AL		Total		P-value
	(N = 7,194)		(N = 113,051)		(N = 120,245)
Age, (years)							0.1347
Mean ± SD	63.89	± 11.67	64.10	± 11.43	64.09	± 11.45
Median (IQR)	64.00	(56.00, 73.00)	65.00	(56.00, 73.00)	65.00	(56.00, 73.00)
Gender, n(%)							<0.0001
Male	4,797	(66.68)	70,553	(62.41)	75,350	(62.66)
Female	2,397	(33.32)	42,498	(37.59)	44,895	(37.34)
Primary diagnosis							<0.0001
C18 (colon)	2,801	(38.94)	51,587	(45.63)	54,388	(45.23)
C19 (colorectal junction)	1,007	(14.00)	17,917	(15.85)	18,924	(15.74)
C20 (rectum)	3,386	(47.06)	43,547	(38.52)	46,933	(39.03)
CCI Score
Mean ± SD	3.75	± 2.90	3.18	± 2.43	3.21	± 2.46	<0.0001
Median	3.00	(2.00, 5.00)	2.00	(2.00, 4.00)	3.00	(2.00, 4.00)
Comorbidities at baseline
Diabetes	1,761	(24.48)	23,315	(20.62)	25,076	(20.85)	<0.0001
COPD	516	(7.17)	7,156	(6.33)	7,672	(6.38)	0.0046
Metastatic disease	910	(12.65)	8,372	(7.41)	9,282	(7.72)	<0.0001
Ischemic heart disease	733	(10.19)	9,292	(8.22)	10,025	(8.34)	<0.0001
Ischemic stroke	385	(5.35)	4,488	(3.97)	4,873	(4.05)	<0.0001
Radiation therapy	922	(12.82)	9,845	(8.71)	10,767	(8.95)	<0.0001
Surgical procedure							<0.0001
AR	2,563	(35.63)	50,669	(44.82)	53,232	(44.27)
LAR	4,420	(61.44)	60,560	(53.57)	64,980	(54.04)
uLAR	211	(2.93)	1,822	(1.61)	2,033	(1.69)
Surgical approach							<0.0001
Open	2,338	(32.50)	28,789	(25.47)	31,127	(25.89)
Laparoscopic	4,856	(67.50)	84,262	(74.53)	89,118	(74.11)
Protective ostomy	1,953	(27.15)	17,894	(15.83)	19,847	(16.51)	<0.0001
Multiple circular stapler use	78	(1.08)	863	(0.76)	941	(0.78)	0.0027
Multiple linear stapler use (≥3)	4,160	(57.83)	55,327	(48.94)	59,487	(49.47)	<0.0001
Year of surgery							<0.0001
2008~2011	2,148	(29.86)	35,398	(31.31)	37,546	(31.22)
2012~2015	2,355	(32.74)	38,605	(34.15)	40,960	(34.06)
2016~2020	2,691	(37.41)	39,048	(34.54)	41,739	(34.71)
Number of beds							0.1745
0~499 beds	995	(13.83)	16,330	(14.44)	17,325	(14.41)
500~999 beds	3,692	(51.32)	56,863	(50.30)	60,555	(50.36)
1000 or more	2,507	(34.85)	39,858	(35.26)	42,365	(35.23)
Blood transfusion (within 30 days)	2,334	(32.44)	19,219	(17.00)	21,553	(17.92)	<0.0001
Urinary tract injury	18	(0.25)	86	(0.08)	104	(0.09)	<0.0001
Pulmonary embolism	45	(0.63)	152	(0.13)	197	(0.16)	<0.0001
Acute renal failure	190	(2.64)	254	(0.22)	444	(0.37)	<0.0001

AL = anastomotic leakage; CCI = Charlson comorbidity index; SD = standard deviation; COPD = chronic obstructive pulmonary disease; AR = anterior resection; LAR = low anterior resection; uLAR = ultra-low anterior resection

Data is presented as the number of patients (%) except for age and CCI score and the denominator of % is the total number of patients in each category.

Risk factors for anastomotic leakage

The results from logistic regression predicting the odds of AL are shown in Table 3. Most of the variables had an association with AL in univariable analysis, except for age and the number of hospital beds. In multivariable analysis, male gender had the increased odds of AL (odds ratio [OR]: 1.176, 95% confidence interval [CI]: 1.117 to 1.239), and older age was associated with the reduced odds of AL (OR 0.995, 95% CI, 0.993 to 0.997). A primary diagnosis of C19 (rectosigmoid junction cancer) decreased the odds of AL (OR 0.888, 95% CI, 0.821 to 0.96, reference: C20 (rectal cancer)), but C18 (colon cancer) did not have a statistically significant association with AL. Diabetes (OR 1.124, 95% CI, 1.059 to 1.192), metastatic disease (OR 1.44, 95% CI, 1.335 to 1.553), ischemic heart disease (OR 1.156, 95% CI, 1.064 to 1.257), and ischemic stroke (OR 1.237, 95% CI 1.107 to 1.382) each increased the odds of AL. Among operative characteristics, protective ostomy (OR 1.546, 95% CI 1.448 to 1.65) and multiple linear stapler use (OR 1.192, 95% CI 1.133 to 1.255) were associated with the higher odds of AL. Meanwhile, AR and LAR were associated with the lower odds of AL versus uLAR (OR 0.619, 95% CI 0.524 to 0.731 and OR 0.691, 95% CI 0.594 to 0.805), and the laparoscopic approach was also associated with the lower odds of AL (OR 0.849, 95% CI 0.804 to 0.897).

Table 3

Risk factors and clinical parameters associated with the anastomotic leak from logistic regression.

Variable		Univariable		Multivariable
		Odds Ratio (95% CI)	p-value	Odds Ratio (95% CI)	p-value
Age, (years)		0.998 (0.996, 1.000)	0.1268	0.995 (0.993, 0.997)	<.0001
Gender	Male vs Female	1.205 (1.146, 1.268)	<.0001	1.176 (1.117, 1.239)	<.0001
Primary diagnosis	C18 vs C20	0.698 (0.663, 0.735)	<.0001	0.936 (0.868, 1.009)	0.0851
	C19 vs C20	0.723 (0.672, 0.777)	<.0001	0.888 (0.821, 0.960)	0.0028
CCI Score		1.086 (1.077, 1.096)	<.0001	-
Diabetes (at baseline)	Yes vs No	1.248 (1.180, 1.319)	<.0001	1.124 (1.059, 1.192)	0.0001
COPD (at baseline)	Yes vs No	1.143 (1.042, 1.255)	0.0046	1.087 (0.988, 1.195)	0.0869
Metastatic disease (at baseline)	Yes vs No	1.811 (1.683, 1.948)	<.0001	1.44 (1.335, 1.553)	<.0001
Ischemic heart disease (at baseline)	Yes vs No	1.267 (1.171, 1.372)	<.0001	1.156 (1.064, 1.257)	0.0007
Ischemic stroke (at baseline)	Yes vs No	1.369 (1.230, 1.523)	<.0001	1.237 (1.107, 1.382)	0.0002
Radiation therapy	Yes vs No	1.541 (1.434, 1.656)	<.0001	1.014 (0.933, 1.103)	0.7432
Surgical procedure	AR vs uLAR	0.437 (0.377, 0.506)	<.0001	0.619 (0.524, 0.731)	<.0001
	LAR vs uLAR	0.630 (0.545, 0.729)	<.0001	0.691 (0.594, 0.805)	<.0001
Surgical approach	Laparoscopic vs Open	0.709 (0.674, 0.747)	<.0001	0.849 (0.804, 0.897)	<.0001
Protective ostomy	Yes vs No	1.982 (1.877, 2.092)	<.0001	1.546 (1.448, 1.650)	<.0001
Multiple circular stapler use	Yes vs No	1.427 (1.131, 1.801)	0.0028	1.196 (0.944, 1.514)	0.1377
Multiple linear stapler use (≥3)	Yes vs No	1.430 (1.363, 1.501)	<.0001	1.192 (1.133, 1.255)	<.0001
Number of beds	0~499 vs 1000 or more	0.969 (0.898, 1.045)	0.4104	-
	500~999 vs 1000 or more	1.032 (0.980, 1.088)	0.2341	-
Blood transfusion		2.345 (2.227, 2.469)	<.0001	2.039 (1.928, 2.156)	<.0001
Urinary tract injury		3.295 (1.981, 5.479)	<.0001	2.606 (1.549, 4.384)	0.0003

The covariates of multiple model were selected based on a significance level of 0.15 in the univariable model.

CCI and comorbidities (diabetes, COPD, metastatic disease, ischemic heart disease, ischemic stroke) had high multicollinearity, and the multiple model with comorbidities showed better explanatory power than that with CCI.

We compared the total healthcare costs (the index hospitalization alone, the period from the index date to last follow-up date, and readmissions) by procedure and approach (Table 4) (Fig 2). All cost categories showed the highest costs in patients who received uLAR compared to AR and LAR. For the index hospitalization, uLAR with an open approach had the highest mean cost (11,212 ± 6,198 USD), and AR with a laparoscopic approach had the lowest mean cost (6,689 ± 3,256 USD). While open approaches were more costly than laparoscopic approaches for the index hospitalization for both AR and uLAR, an open approach for LAR was associated with lower index hospitalization costs than the laparoscopic approach.

Table 4

Healthcare costs by procedure & approach.

Variable		Surgical procedure			p-value	p-value by post hoc comparison*
		AR	LAR	uLAR	by KW test	AR vs. LAR	AR vs. uLAR	LAR vs. uLAR
Total costs for the index hospitalization (USD)					<.0001
Open	N	11,534	19,332	261		<.0001	<.0001	<.0001
	Mean ± SD	6,726 ± 4,409	7,069 ± 4,598	11,212 ± 6,198
	Median (IQR)	5,593 (4,257–7,756)	5,969 (4,678–7,951)	9,324 (8,127–12,636)
Laparoscopic	N	41,698	45,648	1,772		<.0001	<.0001	<.0001
	Mean ± SD	6,689 ± 3,256	7,842 ± 3,719	10,893 ± 3,574
	Median (IQR)	6,124 (5,047–7,434)	7,121 (5,863–8,693)	9,986 (8,944–11,676)
p-value by post hoc comparison*		<.0001	<.0001	0.0251
Total cost for the period from index date to last f/u date (USD)					<.0001
Open	N	11,534	19,332	261		<.0001	<.0001	<.0001
	Mean ± SD	7,216 ± 4,535	7,526 ± 4,723	11,954 ± 6,218
	Median (IQR)	6,131 (4,665–8,314)	6,408 (5,028–8,523)	10,077 (8,559–13,902)
Laparoscopic	N	41,698	45,648	1,772		<.0001	<.0001	<.0001
	Mean ± SD	7,151 ± 3,423	8,318 ± 3,869	11,629 ± 4,030
	Median (IQR)	6,559 (5,414–7,971)	7,475 (6,229–9,279)	10,684 (9,438–12,673)
p-value by post hoc comparison*		<.0001	<.0001	0.1957
Total cost for the readmission (USD)					<.0001
Open	N	2,328	3,536	64		0.2464	0.0382	<.0001
	Mean ± SD	1,532 ± 2,174	1,679 ± 2,359	2,137 ± 2,078
	Median (IQR)	1,293 (755–1,917)	1,111 (635–1,848)	1,430 (1,090–2,947)
Laparoscopic	N	8,113	9,216	399		<.0001	<.0001	<.0001
	Mean ± SD	1,618 ± 1,792	1,603 ± 1,910	2,325 ± 2,542
	Median (IQR)	1,344 (811–1,853)	1,142 (639–1,851)	1,579 (970–2,590)
p-value by post hoc comparison*		0.9473	0.9932	0.9972

1 USD = 1,150 Korean won; USD, U.S. dollar

KW = Kruskal-Wallis test; AR = anterior resection; LAR = low anterior resection; uLAR = ultra-low anterior resection; SD = standard deviation; IQR = interquartile range

* Post hoc pairwise multiple comparison analysis was performed by the Dwass, Steel, Critchlow-Fligner method.

Fig 2

Healthcare costs by procedure & approach.

The clipped boxplot trimmed extremely high values based on 20,000,000 KRW. (1 U.S. dollor = 1,150 Korean won) (A) The number of clipped patients in open anterior resection = 256; The number of clipped patients in laparoscopic anterior resection = 334; The number of clipped patients in open low anterior resection = 432; (B) The number of clipped patients in laparoscopic low anterior resection = 792; The number of clipped patients in open ultra-low anterior resection = 17; The number of clipped patients in laparoscopic ultra-low anterior resection = 76.

Economic burden of anastomotic leakage

Economic outcomes for patients with AL and without AL are summarized in Table 5. The mean costs for the index hospitalization were significantly higher for patients with AL compared to those without AL (8,991 vs. 7,153 USD; p<0.0001). Including the 30-day follow-up period, the mean costs were 10,971 USD for patients with AL and 7,531 USD for those without AL (p<0.0001), reflective of a marked increase in post-discharge resource use among patients in the AL cohort. Patients with AL also required prolonged LOS for the index hospitalization versus those without AL (16.78 days vs. 14.22 days; p<0.0001). Readmission costs for patients with AL were higher than those for patients without AL (3,160 vs. 1,316 USD; p<0.0001). AL patients also required longer mean LOS for readmissions compared to those without AL (20.83 days vs. 13.93 days; p<0.0001). The S1 Table shows economic outcomes depending on the presence of AIC.

Table 5

Economic outcomes of the presence of anastomotic leakage.

	With AL	Without AL	Total	P-value
	(N = 7,194)	(N = 113,051)	(N = 120,245)
Total costs for the index hospitalization, (USD)				<0.0001
Mean ± SD	8,991 ± 6,415	7,153 ± 3,619	7,263 ± 3,868
Median	7,399	6,457	6,508
Q1, Q3	5,782–9,965	5,213–8,101	5,239–8,186
Total costs for the period from the index date to last F/U date, (USD)				<0.0001
Mean ± SD	10,971 ± 6,977	7,531 ± 3656	7,737 ± 4,018
Median	9,222	6,851	6,950
Q1, Q3	7,138–12,505	5,538–8,565	5,592–8,743
Total costs for readmissions, (USD)				<0.0001
N	5,762	51,260	57,022
Mean ± SD	3,160 ± 3,815	1,316 ± 1,029	1,643 ± 1,987
Median	2,000	1,134	1,241
Q1, Q3	1,117–3,762	655–1,703	705–1,866
LOS for the index hospitalization, (duration)				<0.0001
Mean ± SD	16.78 ± 13.24	14.22 ± 8.36	14.37 ± 8.75
Median	13.00	12.00	12.00
Q1, Q3	10.00–19.00	10.00–16.00	10.00–16.00
LOS for readmissions, (duration)				<0.0001
N	5,762	51,260	57,022
Mean ± SD	20.83 ± 16.61	13.93 ± 10.58	14.63 ± 11.53
Median	17.00	13.00	13.00
Q1, Q3	10.00–27.00	7.00–18.00	7.00–18.00

1 USD = 1,150 Korean won; USD, U.S. dollar

AL = anastomotic leakage; SD = standard deviation.

Discussion

This study demonstrates that AL after surgery for colorectal cancer is associated with increased costs for the index hospitalization and readmissions. this study is one of the largest studies to measure the economic burden of AL after colorectal surgery, and using a nationwide population dataset, it could overcome selection bias inherent to smaller, single institution studies.

Various clinical parameters were significant predictors of AL in our analysis. Male gender, diabetes mellitus, ischemic heart disease, ischemic stroke, uLAR, multiple stapler usage, blood transfusion, and urinary tract injury are reported to be well-known risk factors for a higher rate of AL [20-23]. However, there were several associations that were a matter of controversy. Previous prospective randomized clinical trials for rectal cancer showed no significant difference in the incidence of AL between open and laparoscopic surgeries [24-26]. In contrast, one cohort study revealed an increased rate of AL in laparoscopic surgery versus open surgery (10.8% vs. 3.4%, p = 0.012) in patients with mid and low rectal cancer [27]. This result, however, needs to be interpreted with caution, and the learning curve inexperience in the introduction period of laparoscopic surgery for rectal cancer may affect the higher AL rate. Our study showed that the odds of AL were statistically lower for laparoscopic surgery than for open surgery. Basically, there are inevitably significant differences in patient selection, surgical procedure, and complication risk between the patient group who underwent laparoscopic surgery and the group who underwent open surgery. A decision on whether to select laparoscopic or open surgery necessarily involves various preoperative evaluations and surgeon’s concerns about anastomotic leakage risks. In addition, the selection of a surgical method can play a role in preventing the occurrence of anastomotic leakage, but it was very difficult to verify this point in a retrospective study using claim data. Another important confounding factor may be a conversion from laparoscopic to open surgery, which could be an indicator of difficulties encountered during surgery. During such conversions anastomosis may be performed manually, which may help to prevent AL. Unfortunately, conversion surgeries could not be ascertained using claims data, so our findings with respect to surgical approach and incidence of AL should be interpreted with caution.

The influence of preoperative chemoradiotherapy (preop-CRT), especially for rectal cancer, on the incidence of AL is also up for discussion. Jang et al. reported that preop-CRT did not have any incremental impact on the occurrence of AL, as the incidence of AL in patients with and without preop-CRT was 7.5% and 8.1%, respectively, after propensity matching (p = 0.781) [28]. Hu et al. reported that neoadjuvant therapy did not statistically increase the incidence of AL (OR 1.16, 95% CI 0.99–1.36; p = 0.07, random effects model) in a meta-analysis [29]. However, other reports revealed that preop-CRT was a significant independent factor for AL [19, 30]. Our large cohort study showed preoperative receipt of radiation therapy to have no statistical association with the odds of AL. Further studies are required to elucidate the true impact of preop-CRT with AL.

With respect to the influence of age on the occurrence of AL, Zaimi et al. reported that increased age every 5 years was protective for AL after colorectal resections in the multivariable analysis (OR 0.965, 95% CI 0.941–0.985, p<0.001) of 45,488 Dutch colorectal cancer patients [31]. Similarly, other population-based studies revealed that old age was associated with a lower risk of AL [20, 32]. In contrast, several studies did not support this or demonstrated that old age was a risk factor for a higher AL rate [19, 33–35]. In this study, we found that age was associated with a lower incidence of AL. Surgical intervention performed by surgeons with caution for elderly patients or healthy survivor effect, indicating that unhealthy patients die before reaching older age might be suggested as possible reasons for less AL in elderly patients [31]. Our study is one of the largest studies that demonstrated the protective effect of age on AL in Asian groups. Although our study did not focus on the impact of age on AL, this observation might be helpful in planning adequate surgical procedures for elderly patients.

A previous study found short-term outcomes to be dependent on hospital caseload and a degree of specialization [36]. Zheng et al. demonstrated this hospital-center effect with respect to hospital LOS and in-hospital mortality for patients with stage I-III colon cancer who were treated with laparoscopic colectomies [37]. Among the 120,245 patients who were enrolled in our study, only 17,325 (14.41%) underwent surgeries in hospitals with a small number of beds (less than 500 beds). This phenomenon might be attributed to the concentration of hospitals selected by patients in Korea, as more patients from rural areas tend to travel to large metropolitan hospitals for surgery. We did not detect differences in the odds of AL as a function of the hospital size (over 1000 beds, 500–1000 beds, and less than 500 beds). Therefore, surgical processes and/or population risks may not vary across hospitals within these size categories. Curative resection for CRC is regarded as a complex procedure which requires highly trained and experienced surgeons. It was not possible to conduct the analysis of this factor, since the source data lack information on hospital and surgeon caseload.

Hospital LOS was longer among AL patients than those without AL in our study, though the gap between the two groups was relatively small (mean 16.78 days in the AL group vs. 14.22 days in the no AL group, p<0.001). In contrast, Lee et al. observed a much larger incremental difference in the USA (12 days in the AL group vs. 5 days in the no AL group, p<0.0001) [17]. In Korea, many patients may not seek early discharge, given that hospital room charges are reimbursed by the national insurance system. The impact of AL on hospital LOS may therefore depend on historical practices and incentives within local healthcare systems. Despite modest differences in LOS, the total costs for the index hospitalization were significantly higher in patients with AL, so were the total costs for readmissions; these observations corroborate findings from a previous study [17].

One of the critical limitations of this study is that the presence of AL was not identified using individual patients’ medical records. The characteristics of claim data archived by the Korean Health Insurance Review and Assessment Service (HIRA) are not suitable to find a definitive case of AL. There is a precedent, however, of using surrogate indicators of AL from similarly structured data sources. Ashraf et al. reported the economic impact of AL using 23,388 patients registered at the Hospital Episode Statistics (HES) dataset from various National Health Service (NHS) hospitals in England [38]. In that study, researchers defined the AL event as “relaparotomy within 28 days of surgery” because no valid diagnostic code for AL was available [38]. In a recent study using a large claims dataset in the United States, Lee et al. inferred the occurrence of AL from the presence of an abscess, septicemia, peritonitis, or infection among 239,350 patients undergoing colorectal surgery [17]. As with these prior studies, we selected possible clinical patterns representing ALs according an “operational definition”. Nonetheless, we could not make definitive conclusions about the presence of AL, leading us to define additional interventional cases (AIC) in order to overcome the shortcomings. Repeated analyses using the definition of AIC confirmed that the relevant indicators and economic aspects were quite similar. However, as a prior study noted, not all of the relaparotomy within 28 days was done due to AL only [38]; in the case of our study, surgical relaparotomy and radiologic intervention were not always performed on account of AL only.

Another major limitation of this study relates to a lack of cancer staging data. Surgical extent, combined resection, and receipt of chemotherapy are largely dependent on cancer stage. Thus, a stage-specific comparison would be required to eliminate this confounding factor. Since claims data lack staging information, further research with a data source inclusive of surgical morbidity, leakage status, and cancer stage is required.

In conclusion, this research demonstrated AL to be associated with a significantly increased economic burden in the nationwide dataset. Despite limitations inherent to the reliance on surrogate endpoints to find AL cases, this work underscores an opportunity to reduce the economic burden associated with treatments required for AL.

Economic outcomes of the presence of Additional Intervention Cases(AIC).

(DOCX)

Click here for additional data file.

18 Nov 2021

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Alberto Meyer, MD, PhD

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1.Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for stating the following financial disclosure:

(This work was supported by Johnson & Johnson Medical Korea(https://www.jnjmedicaldevices.com/ko-KR). HK and HJP, employees of the sponsor, participated in study design and preparation of the manuscript. BL who was paid for the statistical analysis from the sponsor.).

3. We note that one or more of the authors is affiliated with the funding organization, indicating the funder may have had some role in the design, data collection, analysis or preparation of your manuscript for publication; in other words, the funder played an indirect role through the participation of the co-authors. If the funding organization did not play a role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript and only provided financial support in the form of authors' salaries and/or research materials, please do the following:

A. Review your statements relating to the author contributions, and ensure you have specifically and accurately indicated the role(s) that these authors had in your study. These amendments should be made in the online form.

B. Confirm in your cover letter that you agree with the following statement, and we will change the online submission form on your behalf:

“The funder provided support in the form of salaries for authors [insert relevant initials], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.

C. Please note that in order to use the direct billing option the corresponding author must be affiliated with the chosen institute. Please either amend your manuscript to change the affiliation or corresponding author, or email us at plosone@plos.org with a request to remove this option.

4.Thank you for stating the following in the Competing Interests section:

(I have read the journal's policy and the authors of this manuscript have the following competing interests: HK and HJP are employees of Johnson & Johnson Medical Korea. BL is an employee of RexSoft, which was paid for the statistical analysis from Johnson & Johnson Medical Korea.).

Please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials, by including the following statement: ""This does not alter our adherence to  PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests).  If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your updated Competing Interests statement in your cover letter; we will change the online submission form on your behalf.

5. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

""Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

6. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ

Additional Editor Comments:

Thank you for submitting your manuscript to PLOS ONE.

After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands.

Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process. As you can see, the article needs improvement, especially, in the light of methodology of how the findings compare colon to rectal cancer and surgery with neoadjuvant treatment.

We look forward to receiving your revised manuscript.

King regards,

Alberto Meyer

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The English language needs to be polished for correcting some writing and alphabetical errors thorough text.

It is suggested that statistical analyses are reviewed/re-checked by a statistician.

The statistical values and numbers are suggested to be reviewed again to confirm by authors.

Reviewer #2: The authors retrospectively analyzed the risk factors and economic burden of anastomatic leakage (AL) after colorectal cancer. The study included large samples from Korean patient database, the analysis method was proper, and the manuscript was well organized. However, the limitations of the study were also obvious. There are some important issues to be further addressed before acceptance:

1. The study include rectal cancer and sigmoid colon cancer, which were highly heterogeneous in surgery and the AL risk. Why did the author mixed such two kinds of disease together?

2. In the method part, the author clarified that "all patients underwent surgery as the primary treatment", that means all the included patients did not receive neoadjuvant chemoradiotherapy. So the conclusion of this study is not representative to all the CRC patients who got surgery, but only to those who got surgery first (without neoadjuvant chemoradiotherapy).

3. The AL has classification criteria of AL (grade), which was helpful to evaluate the severity of AL, and also helpful to explain the cost and LOS data, but the authors did not show the grade data.

4. Due to the big difference in therapy, the result has its own bias and limitation. For example, the open surgery and laparoscopy surgery has huge diffence in patient selection, surgical procedure and complication risk; so the authors need fully discuss the limitation of the results.

Reviewer #3: Authors conducted a retrospective, nationwide research about the clinical and economic burden caused by anastomotic leakage (AL) in Korea. Of 156,545 patients undergoing anterior resection (AR), low anterior resection (LAR), or ultra-low anterior resection (uLAR) for colorectal cancer (CRC) between January 1, 2007 and January 31, 2020 were included. Among 120,245 patients who met the eligibility criteria, 7,194 (5.98%) patients had AL within 30 days after surgery. Male gender, comorbidities, protective ostomy, and multiple linear stapler use were associated with a higher odds of AL. Older age, rectosigmoid junction cancer, AR, LAR, and laparoscopic approach were associated with reduced odds of AL. Patients with AL incurred higher costs for index hospitalization compared to those without AL (8,991 vs. 7,153 USD; p <0.0001). Patients with AL also required longer LOS (16.78 vs. 14.22 days; p <0.0001) and readmissions (20.83 vs. 13.93 days; p <0.0001). In summary, they concluded that patients requiring resection for CRC, the occurrence of AL was associated with significantly increased costs and LOS. Preventing AL could not only provide for superior clinical outcomes, but also reduce the economic burden for patients and payers. The results seems interesting and appealing; however, there are a lot of criticisms and have several issues that the authors need to address before the manuscript is suitable for publication.

Major Compulsory Revisions:

1. Clinical outcomes paragraph. The following variables were identified as indicators of surgical complications within 30 days after index surgery: AL, infection, blood transfusion, urinary tract injury, ileus, pneumonia, pulmonary embolism, acute renal failure. Blood transfusion is defined as surgical complications? The transfusion units should be considered. Acute myocardial infarction, wound infection and stroke should be also included as surgical complications.

2. Since there is no specific diagnosis code for AL, presence of the following procedures was required when AL occurrence during in-hospital care comprised operational AL definitions: (1) Imaging study including computed tomography scans (2) Administration of antibacterial drugs (more than 7 consecutive days after the surgery), the above two procedures were hard to be defined as AL. In addition, the nonsynchronous creation of colostomy after AR, LAR, and laparoscopic approach should be considered as AL.

3. In Table 2: Demographics and Perioperative Clinical characteristics. How did authors could identify some variables were surgical complications or risk factors of AL? For example, ischemic heart disease, ischemic stroke, etc.

4. Table 3: Risk factors of anastomotic leak from logistic regression. Age, (years) was an independent variable by multivariate analysis but not by univariate analysis? The subgroup analysis of CCI Score should be mentioned here, and no multivariate analysis for CCI score? Protective ostomy is often performed in ultra-low anterior resection (uLAR), of which might be considered as a compounding factor but not as a risk factor. Furthermore, robotic-assisted surgery vs laparoscopy vs open surgery is suggested to be analyzed if this procedure is related to AL?

5. Table 4. Healthcare costs by procedures & approaches. Only open vs laparoscopic approaches? How about in the comparison with robotic-assisted surgery? The relevant information regarding robotic-assisted surgery is important in recent years.

6. Table 5. Economic outcomes by the presence of anastomotic leakage. Mean LOS and Median LOS for index hospitalization, (duration) was 14.22 ± 8.36 and 12 days, respectively. In fact, it was relatively longer compared to Western countries and even longer than some Asian countries.

7. If authors could use Health Insurance Review and Assessment Service (HIRA) archives claim data to analyze the difference of overall survival between AL vs non-AL patients?

8. In Abstract section: Male gender, comorbidities, protective ostomy, and multiple linear stapler use were associated with a higher odds of AL. The above statement should be amended according to complete results in Table 3.

Minor Essential Revisions:

1. Please correct the typos and grammatical error by English-editing with the certificate enclosed.

2. Abbreviations in the Tables must be shown their corresponding full name in the footnotes.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

31 Mar 2022

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

As you can see, the article needs improvement, especially, in the light of methodology of how the findings compare colon to rectal cancer and surgery with neoadjuvant treatment.

Please submit your revised manuscript by 01/02/2022. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

• A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

• An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1.Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for stating the following financial disclosure:

(This work was supported by Johnson & Johnson Medical Korea(https://www.jnjmedicaldevices.com/ko-KR). HK and HJP, employees of the sponsor, participated in study design and preparation of the manuscript. BL who was paid for the statistical analysis from the sponsor.).

3. We note that one or more of the authors is affiliated with the funding organization, indicating the funder may have had some role in the design, data collection, analysis or preparation of your manuscript for publication; in other words, the funder played an indirect role through the participation of the co-authors. If the funding organization did not play a role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript and only provided financial support in the form of authors' salaries and/or research materials, please do the following:

A. Review your statements relating to the author contributions, and ensure you have specifically and accurately indicated the role(s) that these authors had in your study. These amendments should be made in the online form.

Response: Thank you for your comments. We reviewed the author contribution and modified it. This statement was applied in the revised manuscript and online form.

B. Confirm in your cover letter that you agree with the following statement, and we will change the online submission form on your behalf:

“The funder provided support in the form of salaries for authors [insert relevant initials], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.

Response: Thank you for your comments. We agree with your suggestion to revise the statement because the funders did not have any additional role in the study and manuscript.

C. Please note that in order to use the direct billing option the corresponding author must be affiliated with the chosen institute. Please either amend your manuscript to change the affiliation or corresponding author, or email us at plosone@plos.org with a request to remove this option.

4.Thank you for stating the following in the Competing Interests section:

(I have read the journal's policy and the authors of this manuscript have the following competing interests: HK and HJP are employees of Johnson & Johnson Medical Korea. BL is an employee of RexSoft, which was paid for the statistical analysis from Johnson & Johnson Medical Korea.).

Please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials, by including the following statement: ""This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests). If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your updated Competing Interests statement in your cover letter; we will change the online submission form on your behalf.

Response: We used the claim data generated in the Health Insurance Review and Assessment Service (HIRA). HIRA has strict regulations to provide the data for research. We described the legal restrictions to sharing the minimal data set in the cover letter.

5. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

""Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

Response: We used the claim data generated in the Health Insurance Review and Assessment Service (HIRA). HIRA has strict regulations to provide the data for research. We described the legal restrictions to sharing the minimal data set in the cover letter.

Under the public, single-payer system, the claim data were generated in the Health Insurance Review and Assessment Service (HIRA) archives in the process of reimbursing healthcare providers, covering all Korean citizens. In order to use the big data of HIRA, a researcher should submit the application with a study proposal. HIRA review the application for approval of the access to data. Once the approval is given, HIRA retrieves the data from the data warehouse system, which is then uploaded to the system of HIRA. The system is accessible only to the researcher for the study through the designated computer of the HIRA datacenter or authorized remote access only for a limited period. As taking the raw data out is forbidden, the researcher can take out only the analysis results. Therefore, we cannot share the minimal data set of this study. The application process and documents are specified in the Healthcare Bigdata Hub homepage of HIRA.

HIRA homepage address: https://opendata.hira.or.kr/or/orb/useGdInfo.do

6. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ

Response: Thank you for your detailed explanation.

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

________________________________________

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: Yes

Reviewer #3: Yes

________________________________________

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

________________________________________

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The English language needs to be polished for correcting some writing and alphabetical errors thorough text.

Response: Thank you for your comments. We made corrections to the manuscript accordingly.

It is suggested that statistical analyses are reviewed/re-checked by a statistician.

Response: As you commented, the statistician reviewed/re-checked the overall statistical analysis parts.

The statistical values and numbers are suggested to be reviewed again to confirm by authors.

Response: We reviewed all the statistical values and numbers again.

Reviewer #2: The authors retrospectively analyzed the risk factors and economic burden of anastomatic leakage (AL) after colorectal cancer. The study included large samples from Korean patient database, the analysis method was proper, and the manuscript was well organized. However, the limitations of the study were also obvious. There are some important issues to be further addressed before acceptance:

1. The study include rectal cancer and sigmoid colon cancer, which were highly heterogeneous in surgery and the AL risk. Why did the author mixed such two kinds of disease together?

Response: Basically, we agree that colon and rectal cancer have different anastomotic leakage rates. Therefore, there have been many previous studies where the analyses were performed separately. Our study was conducted with the intention of identifying the trend of anastomotic leakage for all colorectal cancer patients nationwide. As you pointed out, we analyzed the differences in anastomotic leakage rates by location such as the colon and the rectum as shown in Table 2. We also presented the healthcare costs by tumor location in Table 4. Thank you for your comments.

2. In the method part, the author clarified that "all patients underwent surgery as the primary treatment", that means all the included patients did not receive neoadjuvant chemoradiotherapy. So the conclusion of this study is not representative to all the CRC patients who got surgery, but only to those who got surgery first (without neoadjuvant chemoradiotherapy).

Response: Thank you for your comments. The sentence "all patients underwent surgery as the primary treatment” was meant to include all the patients who underwent surgeries. As you pointed out, however, this may cause misunderstandings about patients who received neoadjuvant treatment. Therefore, it would be better to modify the above expression to "all patients underwent surgery as the principal treatment" as follows. Thank you for your comments.

3. The AL has classification criteria of AL (grade), which was helpful to evaluate the severity of AL, and also helpful to explain the cost and LOS data, but the authors did not show the grade data.

Response: As you commented, hospital costs and length of stay vary depending on the grade of AL, so if we can know this accurately, we can perform more detailed analysis. However, it was difficult to obtain data on the grades of AL due to the nature of the claim data, so we could not include it in our analysis. Thank you for your comments.

4. Due to the big difference in therapy, the result has its own bias and limitation. For example, the open surgery and laparoscopy surgery has huge diffence in patient selection, surgical procedure and complication risk; so the authors need fully discuss the limitation of the results.

Response: Thank you for your comments. We agree with your comments. We added the paragraph below in the Discussion section

Included in the Discussion section

Basically, there are inevitably significant differences in patient selection, surgical procedure, and complication risk between the patient group who underwent laparoscopic surgery and the group who underwent open surgery. A decision on whether to select laparoscopic or open surgery necessarily involves various preoperative evaluations and surgeon's concerns about anastomotic leakage risks. In addition, the selection of a surgical method can play a role in preventing the occurrence of anastomotic leakage, but it was very difficult to verify this point in a retrospective study using claim data.

Reviewer #3: Authors conducted a retrospective, nationwide research about the clinical and economic burden caused by anastomotic leakage (AL) in Korea. Of 156,545 patients undergoing anterior resection (AR), low anterior resection (LAR), or ultra-low anterior resection (uLAR) for colorectal cancer (CRC) between January 1, 2007 and January 31, 2020 were included. Among 120,245 patients who met the eligibility criteria, 7,194 (5.98%) patients had AL within 30 days after surgery. Male gender, comorbidities, protective ostomy, and multiple linear stapler use were associated with a higher odds of AL. Older age, rectosigmoid junction cancer, AR, LAR, and laparoscopic approach were associated with reduced odds of AL. Patients with AL incurred higher costs for index hospitalization compared to those without AL (8,991 vs. 7,153 USD; p <0.0001). Patients with AL also required longer LOS (16.78 vs. 14.22 days; p <0.0001) and readmissions (20.83 vs. 13.93 days; p <0.0001). In summary, they concluded that patients requiring resection for CRC, the occurrence of AL was associated with significantly increased costs and LOS. Preventing AL could not only provide for superior clinical outcomes, but also reduce the economic burden for patients and payers. The results seems interesting and appealing; however, there are a lot of criticisms and have several issues that the authors need to address before the manuscript is suitable for publication.

Major Compulsory Revisions:

1. Clinical outcomes paragraph. The following variables were identified as indicators of surgical complications within 30 days after index surgery: AL, infection, blood transfusion, urinary tract injury, ileus, pneumonia, pulmonary embolism, acute renal failure. Blood transfusion is defined as surgical complications? The transfusion units should be considered. Acute myocardial infarction, wound infection and stroke should be also included as surgical complications.

Response: Thank you for your comments. Blood transfusion was included as one of the variables in Table 3. Since the situation requiring blood transfusion may reflect the difficulty of surgery, we checked the relationship between blood transfusion and anastomotic leakage. However, like your opinion, there are parts that it is not easy to refer to as risk factors directly, so we have changed the title of Table 3 into "Risk factors and clinical parameters associated with the anastomotic leak from logistic regression."

However, it was very difficult to include the transfusion volume, acute myocardial infarction, wound infection and stroke as covariates due to the characteristics of claim data. This would be one of the main limitations of our study.

2. Since there is no specific diagnosis code for AL, presence of the following procedures was required when AL occurrence during in-hospital care comprised operational AL definitions: (1) Imaging study including computed tomography scans (2) Administration of antibacterial drugs (more than 7 consecutive days after the surgery), the above two procedures were hard to be defined as AL. In addition, the nonsynchronous creation of colostomy after AR, LAR, and laparoscopic approach should be considered as AL.

Response: We agree with your comments. We believe that abdominal CT is performed again during the period of discharge after surgery when it is really necessary to identify a certain serious situation in the abdomen. Therefore, we determined that such examination after colorectal cancer surgery would be required mostly when complications such as leakage are suspected. In addition, if antibiotics are used for more than a week, in Korea, the use of antibiotics is usually limited to 1-2 days after elective colorectal cancer surgery, according to the evaluation criteria of the National Review and Assessment Service. This standard was published as an index that most hospitals satisfy. Therefore, it can be judged that the clinical situation in which antibiotics are used continuously for one week or more after surgery is an emergency in which the use of antibiotics is essential in light of these indicators. For this reason, two items were included in the operational definition. Nevertheless, as we described the limitations in defining the accurate anastomotic leakage using nationwide claim data, it was impossible to collect the correct cases only.

Because nonsynchronous creation of colostomy could not be identified in the claim database, it was impossible to include it as an independent variable. However, we believe that our operational definitions could cover most of the colostomy or ileostomy formation cases because those patients usually underwent APCT again or antibiotics were prescribed longer than usual. Thank you for your comments.

3. In Table 2: Demographics and Perioperative Clinical characteristics. How did authors could identify some variables were surgical complications or risk factors of AL? For example, ischemic heart disease, ischemic stroke, etc.

Response: We agree with your opinion. In Table 3, ischemic heart disease and ischemic stroke were not surgical complications. These variables were identified in each patient within 1 year before surgery. Thank you for your comments.

4. Table 3: Risk factors of anastomotic leak from logistic regression. Age, (years) was an independent variable by multivariate analysis but not by univariate analysis? The subgroup analysis of CCI Score should be mentioned here, and no multivariate analysis for CCI score? Protective ostomy is often performed in ultra-low anterior resection (uLAR), of which might be considered as a compounding factor but not as a risk factor.

Response: Thank you for your comments. We constructed an initial multiple model using variables that were significant based on a significance level of 0.15 in the univariable model. Because of this, in multiple models, the number of beds was excluded.

At this time, CCI and comorbidities (Diabetes, COPD, Metastatic disease, Ischemic heart disease, Ischemic stroke) were variables with strong multicollinearity, and only one of them had to be included in the model. Therefore, the model with higher explanatory power was determined to be the final model by comparing multiple models with comorbidities excluded and multiple models with CCI excluded and comorbidities included.

We described this procedure in the statistical analysis part in the manuscript as follows: The multivariable model included factors that showed the statistical significance based on univariable analysis at the significance level of 0.15, and it was evaluated by generalized variance inflation factor(VIF) for multicollinearity and the final model was chosen based on Akaike Information Criteria(AIC).

Furthermore, robotic-assisted surgery vs laparoscopy vs open surgery is suggested to be analyzed if this procedure is related to AL?

Response: Thank you for your comments. Robotic-assisted surgery for colorectal resection is not covered by insurance in Korea. Since we used the nationwide claim data and robot surgery is not covered by insurance, it was not included in this analysis by default.

5. Table 4. Healthcare costs by procedures & approaches. Only open vs laparoscopic approaches? How about in the comparison with robotic-assisted surgery? The relevant information regarding robotic-assisted surgery is important in recent years.

Response: Thank you for your comments. Robotic-assisted surgery for colorectal resection is not covered by insurance in Korea. Since we used the nationwide claim data and robot surgery is not covered by insurance, it was not included in this analysis by default.

6. Table 5. Economic outcomes by the presence of anastomotic leakage. Mean LOS and Median LOS for index hospitalization, (duration) was 14.22 ± 8.36 and 12 days, respectively. In fact, it was relatively longer compared to Western countries and even longer than some Asian countries.

Response: Thank you for your comments. We agree with your opinion. It is not possible to accurately analyze the cause of hospital stay being considerably longer than in other cases. Although there is a difference in the room rate in Korea, when it is covered by insurance, the cheapest room rate is about 10 dollars a day. Therefore, patients want to stay in the hospital as long as possible, rather than being discharged early. This is thought to be one of the main causes. But, over the last 5-6 years, each hospital has been applying ERAS, etc. to discharge patients as early as possible to improve the management of the hospital, so the length of stay has been shortened. Therefore, it seems that the long-term hospitalization was probably reflected in the past, and a relatively large number of patients who underwent open surgery are also considered one of the reasons.

7. If authors could use Health Insurance Review and Assessment Service (HIRA) archives claim data to analyze the difference of overall survival between AL vs non-AL patients?

Response: As the HIRA database includes only the information on reimbursement, there is no information on death, so we could not identify the mortality. Thank you for your comments.

8. In Abstract section: Male gender, comorbidities, protective ostomy, and multiple linear stapler use were associated with a higher odds of AL. The above statement should be amended according to complete results in Table 3.

Response: Thank you for your comments. As mentioned in the previous section, CCI and the rest of the comorbidities had high multicollinearity, and the multiple model with comorbidities showed better explanatory power than that with CCI. Thus, we just showed the effect size of CCI in the univariable model and excluded that in the final multivariable model.

Included in the revised manuscript

Male gender, comorbidities (diabetes, metastatic disease, ischemic heart disease, ischemic stroke), protective ostomy, and multiple linear stapler use, blood transfusion, and urinary tract injury were associated with the higher odds of AL. Older age, rectosigmoid junction cancer, AR, LAR, and laparoscopic approach were related with the reduced odds of AL.

Minor Essential Revisions:

1. Please correct the typos and grammatical error by English-editing with the certificate enclosed.

Response: Thank you for your comments. We made corrections to the manuscript accordingly.

2. Abbreviations in the Tables must be shown their corresponding full name in the footnotes.

Response: Thank you for your comments. We made corrections to the manuscript accordingly.

________________________________________

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

20 Apr 2022

Risk factors and economic burden of postoperative anastomotic leakage related events in patients who underwent surgeries for colorectal cancer

PONE-D-21-34976R1

Dear Dr. LEE,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Alberto Meyer, MD, PhD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Dear Authors,

Thank you for accepting our recommendations for revision and incorporating the relevant changes.

I am satisfied with your response and thus happy to recommend in favour of publication of your study.

Kind regards

Reviewers' comments:

10 May 2022

PONE-D-21-34976R1

Risk factors and economic burden of postoperative anastomotic leakage related events in patients who underwent surgeries for colorectal cancer

Dear Dr. Lee:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Professor Alberto Meyer

Academic Editor

PLOS ONE