ORIGINAL CONTRIBUTION
Initial Surgical Management of Ulcerative Colitis in the Biologic Era Cristina B. Geltzeiler, M.D.1 • Kim C. Lu, M.D.1 • Brian S. Diggs, Ph.D.1 Karen E. Deveney, M.D.1 • Kian Keyashian, M.D.2 • Daniel O. Herzig, M.D.1 Vassiliki L. Tsikitis, M.D.1 1 Department of Surgery, Oregon Health & Science University, Portland, Oregon 2 Department of Gastroenterology, Oregon Health & Science University, Portland, Oregon
BACKGROUND: The initial minimum operation for ulcerative colitis is a total abdominal colectomy. Healthy patients may undergo proctectomy at the same time; however, for ill patients, proctectomy is delayed. Since the introduction of biologic medications in 2005, ulcerative colitis medical management has changed dramatically. OBJECTIVE: We examined how operative management for ulcerative colitis has changed from the prebiologic to biologic eras. DESIGN: We conducted a retrospective review of data on patients with ulcerative colitis who were included in the Nationwide Inpatient Sample database. SETTINGS: This study was conducted at a single university. PATIENTS: A total of 1,547,852 patients with ulcerative
colitis who were admitted to a US hospital from 1991 to 2011 were included in the study. MAIN OUTCOME MEASURES: We examined patients whose initial operation consisted of total abdominal colectomy without proctectomy versus a total proctocolectomy with or without a pouch. We also examined which operation was done at the time of the construction of an ileoanal pouch. Patients who underwent colectomy and pouch construction in the same hospitalization were compared with those who received pouch formation at a subsequent hospitalization. Financial Disclosure: None reported. Podium presentation at the meeting of The American Society of Colon and Rectal Surgeons, Hollywood, FL, May 17 to 21, 2014. Correspondence: Cristina B. Geltzeiler, M.D., Oregon Health & Science University, Department of Surgery, 3181 SW Sam Jackson Park Rd, Mail Code L223A, Portland, OR 97239. E-mail: [email protected] Dis Colon Rectum 2014; 57: 1358–1363 DOI: 10.1097/DCR.0000000000000236 © The ASCRS 2014
1358
RESULTS: Ulcerative colitis–related admissions rose by 170% during the years examined, and the number of patients who required total abdominal colectomy increased by 44%. Total abdominal colectomy increased by 15%, as opposed to total proctocolectomy (p < 0.001). Pouch construction at a subsequent operation increased by 16% (p = 0.002). Since 2008, total abdominal colectomy has surpassed total proctocolectomy as the most common initial surgical intervention for ulcerative colitis. LIMITATIONS: The Nationwide Inpatient Sample is a retrospective database, and we were limited to examining the variables within it. CONCLUSIONS: Total abdominal colectomy is currently the most common initial operation for patients with ulcerative colitis, and an ileoanal pouch is more frequently constructed at a subsequent hospitalization. These trends coincide with the initiation of biologic treatments and may imply that patients are acutely ill at the time of initial operation. Alternately, there may be surgeon-perceived bias of increased surgical risk or a shift in care to specialized surgeons for pouch construction. KEY WORDS: Biologic era; Ileoanal pouch; Operative management; Total abdominal colectomy; Total proctocolectomy; Ulcerative colitis.
D
espite recent advances in medical therapy for ulcerative colitis (UC), ≈15% to 30% of patients will require surgery.1,2 When surgical treatment of UC is necessary, the initial minimum operation is a total abdominal colectomy (TAC).1 For healthy patients, a proctectomy is performed during the same operation, either as definitive management or as the first stage of a restorative proctocolectomy (ileoanal pouch).3 For ill patients, proctectomy is delayed.1 Since the introduction of biologic medications for the treatment of UC in 2005, it has been Diseases of the Colon & Rectum Volume 57: 12 (2014)
1359
Diseases of the Colon & Rectum Volume 57: 12 (2014)
reported that there is an increased risk for surgical complications in patients who are on biologic medications.4–6 Despite these observations, trends in the operative management of patients with UC in the era of biologic medications are not known. Our hypothesis is that the initial surgical intervention for UC has shifted from total proctocolectomy (TPC) to TAC. Specifically, we hypothesized that the number of TPCs performed has decreased, whereas the number of TACs has increased over time. Using the Nationwide Inpatient Sample (NIS), a population-based database, we examined the number of TACs performed and compared it with the number of TPCs performed as the initial surgical management for UC over a period of 21 years. To confirm our hypothesis that the number of TPCs has decreased over time, we examined whether an ileoanal pouch was constructed at the initial operation or at a subsequent operation from the total colectomy.
An initial operation was defined as an operation that included a TAC. The percentage of those patients who had a TAC only at their initial operation and those who had a TAC with proctectomy (TPC [with or without pouch construction]) were compared over the years and examined. Similarly, the operation performed at the time of pouch construction was compared. The percentage of those patients who had a pouch construction at a subsequent time from their TAC was compared with the percentage who had pouch construction at the time of TAC. Patient demographics for the initial operation and for pouch construction operation were compared. Changes in operative intervention over time were evaluated for statistical differences using linear regression analysis. All of the models and summary statistics incorporated the hierarchical sampling structure of the NIS. All of the analysis was carried out using R version 3.0.2 and the survey package.8,9 This study was reviewed by our institutional review board and determined not to qualify as human subjects research.
MATERIALS AND METHODS
RESULTS
We used the NIS database. The NIS is sustained by the Healthcare Cost and Utilization Project, which, in turn, is sponsored by the Agency for Healthcare Research and Quality. This database consists of the largest all-payer inpatient care database in the United States. It is based on a 20% stratified sample of nonfederal hospital admissions. Each admission is weighted to provide statistical national representation.7 All of the patient admissions with the diagnosis of UC (any subcode of International Classification of Diseases, Ninth Revision (ICD-9; diagnosis code 556) from the years 1991 to 2011 were included. UC-related surgical interventions of interest were identified based on selected ICD-9 diagnosis codes. Operations of interest included those where patients underwent TAC (any subcode of ICD-9 diagnosis code 45.8), TPC (patients who had a code for TAC and a code for proctectomy, defined as 48.40, 48.41, 48.42, 48.43, 48.49, 48.5, 48.50, 48.51, 48.52, 48.59, 48.61, 48.62, 48.63, 48.64, 48.65, or 48.69), ileoanal pouch construction (45.95), pouch construction at a subsequent operation from initial colectomy (defined as patient with code for pouch construction [45.95] without a code for TAC), and those who underwent pouch construction at the time of TAC (those with a code for pouch construction and for TAC). All of the laparoscopic and open procedures were included. We also examined the subset of patients who underwent TAC with a diagnosis of neoplasia (any subcode of ICD-9 diagnosis codes 153 and 154 but excluding 154.8) or carcinoma in situ (ICD-9 diagnosis codes 230.3 and 230.4). Absolute counts of each operation of interest were obtained for each year examined. Trends over time in counts were assessed using a simple weighted linear regression of the counts versus year.
A total of 1,547,852 UC-related admissions were analyzed. Admissions rose from 44,432 in 1991 to 118,611 in 2011, an increase of 170% (p < 0.001; Fig. 1). Demographics for procedures of interest are listed in Table 1. Patients who underwent TPC at their initial operation were similar in age, sex, and race to those who underwent TAC. Patients who underwent pouch construction at the same time as their colectomy were also similar in age, sex, and race to those who underwent pouch construction at a subsequent operation from their colectomy. A total of 78,213 TACs was performed over the study period. The number of patients who required TAC increased from 3398 in 1991 to 4894 in 2011, an increase of 44% (p < 0.001). During the years examined, ileoanal pouch construction occurred in 34,709 patients. This procedure increased in frequency from 1490 patients in 1991 to 1967 patients in 2011, an increase of 32% (p < 0.001; Fig. 1). The number of colectomies with the diagnosis of neoplasia and/or carcinoma in situ was 5498 (7.0% of all colectomies). On examination of each year, this rate ranged from 4.6% to 10.1%, but there was no statistically significant difference when comparing years (p = 0.129). During the 21-year period examined, 33,492 patients had a TAC at their initial operation, whereas 44,721 patients had a TPC with or without pouch construction. As a percentage of all colectomy operations for UC, TAC increased from 41% of all cases in 1991 to 55% of all cases in 2011, whereas, alternately, TPC with or without pouch construction decreased from 59% to only 45% of operations (p < 0.001). Since 2008, TAC surpassed TPC as the most common initial surgical intervention for UC (Fig. 2). Over the study period, there were 14,907 ileoanal pouches constructed during a subsequent operation from
1360
Geltzeiler et al: Surgery for Ulcerative Colitis in the Biologic Era
National estimate of number of cases 125,000
Admissions
100,000
75,000
50,000 7000 6000 5000 TAC
4000 3000 2000
3000
Pouch
2000
1000 FDA approval of biologics for use in UC 1990
1995
2000 Year
2005
2010
FIGURE 1. Number of ulcerative colitis (UC) admissions, total abdominal colectomies (TAC), and pouch constructions for each year examined. Vertical bars at points represent the 95% CIs. Vertical line represents 2005, approval of biologic medications for use in UC. FDA = US Food and Drug Administration.
colectomy and 19,802 ileoanal pouches constructed at the same time as TAC. Pouch construction performed during the same operation as colonic resection decreased from 62% to 46% during the study period, whereas pouch construction performed at a subsequent operation increased from 38% to 54% (p = 0.002). Since 2008, pouch construction during a subsequent operation has surpassed pouch construction at time of colectomy as the more prevalent operation (Fig. 3).
DISCUSSION In this population-based study using the NIS database, we have demonstrated a shift in operative trends for the treatment of UC. There has been an increase in patients undergoing TAC as opposed to TPC as the initial operative
intervention. Similarly, we found an increase in the construction of an ileoanal pouch during a subsequent operation after the initial colectomy. These trends coincide with the initiation of biologic treatments, suggesting either that patients present acutely ill at the time of operation or that surgeons may perceive an increased operative risk with the use of biologic agents. Infliximab, an antitumor necrosis factor-α monoclonal antibody, was approved for use in UC for induction and maintenance therapy in 2005.2 It has been shown to decrease the colectomy rate in the short term, yet its longterm effect is less clear.10–12 It has been suggested that the long-term benefits of biologics as maintenance therapy may be limited because of a lack of long-term efficacy and the development of prohibitive medical adverse effects.11,13 This view has been further supported by the fact that the
1361
Diseases of the Colon & Rectum Volume 57: 12 (2014)
TABLE 1. Patient demographics for initial operation and pouch construction operation Initial operation Variable Age, ya Sexb Male Female Raceb White Black Hispanic Asian or Pacific Islander Native American Other
Pouch construction
TAC (N = 33,492)
TPC with or without pouch (N = 44,721)
Pouch at subsequent operation (N = 14,907)
49 (48–50)
45 (45–46)
36 (35–36)
Pouch at time of TAC (N = 19,802) 38 (37–38)
20,262 (54) 17,188 (46)
27,533 (57) 21,071 (43)
8959 (55) 7437 (45)
11,890 (56) 9250 (44)
22,742 (86) 1333 (5) 1343 (5) 250 (1) 45 (
Initial Surgical Management of Ulcerative Colitis in the Biologic Era Cristina B. Geltzeiler, M.D.1 • Kim C. Lu, M.D.1 • Brian S. Diggs, Ph.D.1 Karen E. Deveney, M.D.1 • Kian Keyashian, M.D.2 • Daniel O. Herzig, M.D.1 Vassiliki L. Tsikitis, M.D.1 1 Department of Surgery, Oregon Health & Science University, Portland, Oregon 2 Department of Gastroenterology, Oregon Health & Science University, Portland, Oregon
BACKGROUND: The initial minimum operation for ulcerative colitis is a total abdominal colectomy. Healthy patients may undergo proctectomy at the same time; however, for ill patients, proctectomy is delayed. Since the introduction of biologic medications in 2005, ulcerative colitis medical management has changed dramatically. OBJECTIVE: We examined how operative management for ulcerative colitis has changed from the prebiologic to biologic eras. DESIGN: We conducted a retrospective review of data on patients with ulcerative colitis who were included in the Nationwide Inpatient Sample database. SETTINGS: This study was conducted at a single university. PATIENTS: A total of 1,547,852 patients with ulcerative
colitis who were admitted to a US hospital from 1991 to 2011 were included in the study. MAIN OUTCOME MEASURES: We examined patients whose initial operation consisted of total abdominal colectomy without proctectomy versus a total proctocolectomy with or without a pouch. We also examined which operation was done at the time of the construction of an ileoanal pouch. Patients who underwent colectomy and pouch construction in the same hospitalization were compared with those who received pouch formation at a subsequent hospitalization. Financial Disclosure: None reported. Podium presentation at the meeting of The American Society of Colon and Rectal Surgeons, Hollywood, FL, May 17 to 21, 2014. Correspondence: Cristina B. Geltzeiler, M.D., Oregon Health & Science University, Department of Surgery, 3181 SW Sam Jackson Park Rd, Mail Code L223A, Portland, OR 97239. E-mail: [email protected] Dis Colon Rectum 2014; 57: 1358–1363 DOI: 10.1097/DCR.0000000000000236 © The ASCRS 2014
1358
RESULTS: Ulcerative colitis–related admissions rose by 170% during the years examined, and the number of patients who required total abdominal colectomy increased by 44%. Total abdominal colectomy increased by 15%, as opposed to total proctocolectomy (p < 0.001). Pouch construction at a subsequent operation increased by 16% (p = 0.002). Since 2008, total abdominal colectomy has surpassed total proctocolectomy as the most common initial surgical intervention for ulcerative colitis. LIMITATIONS: The Nationwide Inpatient Sample is a retrospective database, and we were limited to examining the variables within it. CONCLUSIONS: Total abdominal colectomy is currently the most common initial operation for patients with ulcerative colitis, and an ileoanal pouch is more frequently constructed at a subsequent hospitalization. These trends coincide with the initiation of biologic treatments and may imply that patients are acutely ill at the time of initial operation. Alternately, there may be surgeon-perceived bias of increased surgical risk or a shift in care to specialized surgeons for pouch construction. KEY WORDS: Biologic era; Ileoanal pouch; Operative management; Total abdominal colectomy; Total proctocolectomy; Ulcerative colitis.
D
espite recent advances in medical therapy for ulcerative colitis (UC), ≈15% to 30% of patients will require surgery.1,2 When surgical treatment of UC is necessary, the initial minimum operation is a total abdominal colectomy (TAC).1 For healthy patients, a proctectomy is performed during the same operation, either as definitive management or as the first stage of a restorative proctocolectomy (ileoanal pouch).3 For ill patients, proctectomy is delayed.1 Since the introduction of biologic medications for the treatment of UC in 2005, it has been Diseases of the Colon & Rectum Volume 57: 12 (2014)
1359
Diseases of the Colon & Rectum Volume 57: 12 (2014)
reported that there is an increased risk for surgical complications in patients who are on biologic medications.4–6 Despite these observations, trends in the operative management of patients with UC in the era of biologic medications are not known. Our hypothesis is that the initial surgical intervention for UC has shifted from total proctocolectomy (TPC) to TAC. Specifically, we hypothesized that the number of TPCs performed has decreased, whereas the number of TACs has increased over time. Using the Nationwide Inpatient Sample (NIS), a population-based database, we examined the number of TACs performed and compared it with the number of TPCs performed as the initial surgical management for UC over a period of 21 years. To confirm our hypothesis that the number of TPCs has decreased over time, we examined whether an ileoanal pouch was constructed at the initial operation or at a subsequent operation from the total colectomy.
An initial operation was defined as an operation that included a TAC. The percentage of those patients who had a TAC only at their initial operation and those who had a TAC with proctectomy (TPC [with or without pouch construction]) were compared over the years and examined. Similarly, the operation performed at the time of pouch construction was compared. The percentage of those patients who had a pouch construction at a subsequent time from their TAC was compared with the percentage who had pouch construction at the time of TAC. Patient demographics for the initial operation and for pouch construction operation were compared. Changes in operative intervention over time were evaluated for statistical differences using linear regression analysis. All of the models and summary statistics incorporated the hierarchical sampling structure of the NIS. All of the analysis was carried out using R version 3.0.2 and the survey package.8,9 This study was reviewed by our institutional review board and determined not to qualify as human subjects research.
MATERIALS AND METHODS
RESULTS
We used the NIS database. The NIS is sustained by the Healthcare Cost and Utilization Project, which, in turn, is sponsored by the Agency for Healthcare Research and Quality. This database consists of the largest all-payer inpatient care database in the United States. It is based on a 20% stratified sample of nonfederal hospital admissions. Each admission is weighted to provide statistical national representation.7 All of the patient admissions with the diagnosis of UC (any subcode of International Classification of Diseases, Ninth Revision (ICD-9; diagnosis code 556) from the years 1991 to 2011 were included. UC-related surgical interventions of interest were identified based on selected ICD-9 diagnosis codes. Operations of interest included those where patients underwent TAC (any subcode of ICD-9 diagnosis code 45.8), TPC (patients who had a code for TAC and a code for proctectomy, defined as 48.40, 48.41, 48.42, 48.43, 48.49, 48.5, 48.50, 48.51, 48.52, 48.59, 48.61, 48.62, 48.63, 48.64, 48.65, or 48.69), ileoanal pouch construction (45.95), pouch construction at a subsequent operation from initial colectomy (defined as patient with code for pouch construction [45.95] without a code for TAC), and those who underwent pouch construction at the time of TAC (those with a code for pouch construction and for TAC). All of the laparoscopic and open procedures were included. We also examined the subset of patients who underwent TAC with a diagnosis of neoplasia (any subcode of ICD-9 diagnosis codes 153 and 154 but excluding 154.8) or carcinoma in situ (ICD-9 diagnosis codes 230.3 and 230.4). Absolute counts of each operation of interest were obtained for each year examined. Trends over time in counts were assessed using a simple weighted linear regression of the counts versus year.
A total of 1,547,852 UC-related admissions were analyzed. Admissions rose from 44,432 in 1991 to 118,611 in 2011, an increase of 170% (p < 0.001; Fig. 1). Demographics for procedures of interest are listed in Table 1. Patients who underwent TPC at their initial operation were similar in age, sex, and race to those who underwent TAC. Patients who underwent pouch construction at the same time as their colectomy were also similar in age, sex, and race to those who underwent pouch construction at a subsequent operation from their colectomy. A total of 78,213 TACs was performed over the study period. The number of patients who required TAC increased from 3398 in 1991 to 4894 in 2011, an increase of 44% (p < 0.001). During the years examined, ileoanal pouch construction occurred in 34,709 patients. This procedure increased in frequency from 1490 patients in 1991 to 1967 patients in 2011, an increase of 32% (p < 0.001; Fig. 1). The number of colectomies with the diagnosis of neoplasia and/or carcinoma in situ was 5498 (7.0% of all colectomies). On examination of each year, this rate ranged from 4.6% to 10.1%, but there was no statistically significant difference when comparing years (p = 0.129). During the 21-year period examined, 33,492 patients had a TAC at their initial operation, whereas 44,721 patients had a TPC with or without pouch construction. As a percentage of all colectomy operations for UC, TAC increased from 41% of all cases in 1991 to 55% of all cases in 2011, whereas, alternately, TPC with or without pouch construction decreased from 59% to only 45% of operations (p < 0.001). Since 2008, TAC surpassed TPC as the most common initial surgical intervention for UC (Fig. 2). Over the study period, there were 14,907 ileoanal pouches constructed during a subsequent operation from
1360
Geltzeiler et al: Surgery for Ulcerative Colitis in the Biologic Era
National estimate of number of cases 125,000
Admissions
100,000
75,000
50,000 7000 6000 5000 TAC
4000 3000 2000
3000
Pouch
2000
1000 FDA approval of biologics for use in UC 1990
1995
2000 Year
2005
2010
FIGURE 1. Number of ulcerative colitis (UC) admissions, total abdominal colectomies (TAC), and pouch constructions for each year examined. Vertical bars at points represent the 95% CIs. Vertical line represents 2005, approval of biologic medications for use in UC. FDA = US Food and Drug Administration.
colectomy and 19,802 ileoanal pouches constructed at the same time as TAC. Pouch construction performed during the same operation as colonic resection decreased from 62% to 46% during the study period, whereas pouch construction performed at a subsequent operation increased from 38% to 54% (p = 0.002). Since 2008, pouch construction during a subsequent operation has surpassed pouch construction at time of colectomy as the more prevalent operation (Fig. 3).
DISCUSSION In this population-based study using the NIS database, we have demonstrated a shift in operative trends for the treatment of UC. There has been an increase in patients undergoing TAC as opposed to TPC as the initial operative
intervention. Similarly, we found an increase in the construction of an ileoanal pouch during a subsequent operation after the initial colectomy. These trends coincide with the initiation of biologic treatments, suggesting either that patients present acutely ill at the time of operation or that surgeons may perceive an increased operative risk with the use of biologic agents. Infliximab, an antitumor necrosis factor-α monoclonal antibody, was approved for use in UC for induction and maintenance therapy in 2005.2 It has been shown to decrease the colectomy rate in the short term, yet its longterm effect is less clear.10–12 It has been suggested that the long-term benefits of biologics as maintenance therapy may be limited because of a lack of long-term efficacy and the development of prohibitive medical adverse effects.11,13 This view has been further supported by the fact that the
1361
Diseases of the Colon & Rectum Volume 57: 12 (2014)
TABLE 1. Patient demographics for initial operation and pouch construction operation Initial operation Variable Age, ya Sexb Male Female Raceb White Black Hispanic Asian or Pacific Islander Native American Other
Pouch construction
TAC (N = 33,492)
TPC with or without pouch (N = 44,721)
Pouch at subsequent operation (N = 14,907)
49 (48–50)
45 (45–46)
36 (35–36)
Pouch at time of TAC (N = 19,802) 38 (37–38)
20,262 (54) 17,188 (46)
27,533 (57) 21,071 (43)
8959 (55) 7437 (45)
11,890 (56) 9250 (44)
22,742 (86) 1333 (5) 1343 (5) 250 (1) 45 (
